Masonry
C7. (Systematic Rehabilitation)
C7.1 Scope
The scope of Chapter 7 is limited to masonry elements
that are considered to resist lateral seismic forces as
structural members. The chapter includes walls and
infill panels subjected to in-plane and out-of-plane
forces. Material given is intended to be used directly
with the Analysis Procedures prescribed in Chapter 3.
All other masonry elements are addressed in Chapters 4
and 11.
C7.2 Historical Perspective
C7.2.1 General
Masonry is the oldest of all construction materials,
dating back more than eight millennia to cultures
around the globe. Early masonries consisted of stone
units with no mortar. The structural action in this form
of masonry is much different than that of modern-day
clay-unit and concrete masonry, which is found in
nearly all existing masonry buildings in the United
States, with the exception of some historic buildings
that predate the 1850s.
Most masonry buildings in the United States
constructed before the 20th century consisted of
unreinforced clay-unit masonry. Wythes of brick were
commonly tied with brick headers spaced at every sixth
or seventh course. Because no other construction
material was used for the walls, these building systems
represented the first introduction to engineered masonry
construction, although seismic considerations were
often neglected in the design. Early mortars consisted of
no more than lime and sand, which made the shear and
tensile strength of the masonry quite weak. In the same
era, clay-unit masonry was also used extensively for
infills and cladding on steel frame buildings. Though
the structural properties of the masonry were ignored in
favor of the strong but flexible steel frames,
considerable lateral-force resistance was provided by
the stiff but brittle masonry, as evidenced by substantial
cracking when subjected to earthquake motions.
Following the 1933 Long Beach earthquake,
unreinforced masonry (URM) was banned in
California, giving rise to reinforced masonry (RM)
construction. Today, buildings approaching thirty
FEMA 274 Seismic Rehabilitation Commentary
stories are constructed with stiff, strong, and ductile RM
walls designed with limit states concepts. Both hollow-
clay and concrete block construction have competed
with reinforced concrete and structural steel for the
design of commercial, residential, and industrial
buildings. In addition, clay-unit masonry remains as the
most prevalent material for cladding and veneer on all
types of buildings.
In this section, a short treatise on the history of masonry
materials is presented to educate the user of these
guidelines. Historical summaries are given for:
• clay units
• structural clay tile
• concrete masonry units
• mortar
• reinforced masonry
C7.2.2 Clay Units
Although brick was one of the first products that people
manufactured from clay, the era of modern brick began
only when extrusion machines were developed. A few
bricks were being made by machine in 1833, but the
percentage was small until 1870. With the invention of
the extrusion or stiff-mud brick-making machine, some
manufacturers produced brick containing holes or
“cores” running parallel to either the length or the
height dimension of the unit. These cores were
introduced as an aid to uniform drying of the clay and as
a means of reducing the weight of the unit.
The General Assembly of New Jersey passed a law in
1883 to establish brick dimensions at 9-1/2" x 4-1/2" x
2-3/4". In 1889, in the District of Columbia, the
ordinance of October 31, 1820 was still being enforced,
which fixed a minimal size of brick at 9-1/4" x 4-5/8" x
2-1/4".
In 1929, a report prepared by McBurney and Logwell
summarized that 92% of the brick produced in the
United States had flat-wise compressive strengths
averaging 7,246 psi for both hard and salmon brick.
From the distribution data given, approximately 6% of
7-1
 Chapter 7: Masonry (Systematic Rehabilitation)
the production classified as 1,250 to 2,500 psi, 20% as
2,225 to 4,500 psi, and 74% as over 4,500 psi.
Approximately 40% of the production was 8,000 psi or
over in compressive strength.
Solid brick is now defined as a small building unit, solid
or cored not in excess of 25%, commonly in the form of
a rectangular prism, formed from inorganic,
nonmetallic substances, and hardened in its finished
shape by heat or chemical action. Brick is also available
in larger units with cell or core areas up to 60% of the
cross section. Such units are typically used for
placement of both vertical and horizontal
reinforcement. The term “brick,” when used without a
qualifying adjective, is understood to mean such a unit
or a collection of such units made from clay or shale
hardened by heat.
C7.2.3 Structural Clay Tile
Structural clay tile is a machine-made product first
produced in the United States in New Jersey in 1875.
Structural clay tiles are characterized by the fact that
they are hollow units with parallel cells (hollow
spaces). The shape of the unit is controlled by the die
through which the clay column is extruded. The ease
with which different designs could be produced led to
the development of a wide variety of sizes and patterns.
In 1903, the National Fireproofing Corporation of
Pittsburgh published a handbook and catalog by Henry
L. Hinton, illustrating the products of the company and
presenting data for use in the design of segmental and
flat arch floors. This catalog is of historical interest,
particularly because of the large number of unit designs
illustrated. Hundreds of different shapes are shown for
use in the construction of tile floor arches, partitions,
and walls, and for fireproofing columns, beams, and
girders.
Structural clay tile was used extensively during World
War I. With lumber in critically short supply, hollow-
clay tile was largely relied upon for all types of
buildings. Brick and tile were used for the construction
of mobilization structures, war housing, defense plants,
air fields, and buildings at army and navy bases.
In 1950, structural clay tile was classified under the
following types: Structural Clay Load-Bearing Wall
Tile, Structural Clay Non-Load-Bearing Tile (partition,
furring, and fireproofing), Structural Clay Floor Tile,
Structural Clay Facing Tile, and Structural Glazed
Facing Tile.
7-2 Seismic Rehabilitation Commentary
C7.2.4 Concrete Masonry Units
The earliest specification for hollow concrete block was
proposed by the National Association of Cement Users
in January 1908. The NACU was organized in 1904 and
continued under that name until 1913, when it became
known as the American Concrete Institute (ACI).
In 1905, the United States government adopted concrete
block for its hospitals, warehouses, and barracks in the
Panama Canal Zone and the Philippine Islands.
The 1908 specification called for the block in bearing
walls to have an average strength of 1000 psi at 28 days
with a minimum of 700 psi. Air space was limited to
33% and absorption was to average not more than 15%
(with no single block to exceed 22%). Absorption was
to be measured on a block placed in a pan of water at
least 2" deep. Fine aggregate had to pass a 1/4" mesh
sieve; stone or clean-screened gravel was to go through
a 3/4" sieve and be refused on a 1/4" sieve. A 1-3-4
semi-wet mix was recommended for exposed bearing
walls, and a 1-3-5 mix for a wet cast block. Portland
cement mortar was recommended. Transverse,
compressive, and absorption tests were required, along
with freezing and fire tests when necessary, and the
modulus of rupture at 28 days was to average 150. Any
expense attending such tests was to be met by the
manufacturer of the block.
This first standard specification was adopted in 1910.
Two years later, the practice for curing—which until
that time had consisted of sprinkling with water for
seven days—was revised slightly by the addition of a
new method, the use of steam from 100 to 130°C for 48
hours with a subsequent storing of eight days. This
recommended practice was the first mention of high-
pressure steam curing in block specifications.
In 1916, the absorption rate was changed to 10% at the
end of 48 hours. In 1922 came the first specification for
a non-load-bearing unit, with a requirement of 300 psi.
That same year, the following strengths were suggested:
250, 500, 700, and 1200 psi for non-load-bearing, light-
load-bearing, medium-load-bearing, and heavy-load-
bearing walls, respectively. The ACI accepted these
values as tentative in 1923. The absorption time,
however, was shortened from 48 to 24 hours. A similar
table, with the elimination of the light-load-bearing
unit, was accepted as tentative in 1924, and adopted the
following year.
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
By 1928, more than 80 city building codes had been
revised to eliminate practically all of the legal obstacles
to the increased use of concrete block. Public works
construction by state and local governments had
declined steadily until by 1933 it had virtually ceased.
In 1933, several government agencies were set up to
purchase concrete block. In July 1935, the National
Industrial Recovery Act was invalidated by the U.S.
Supreme Court, but it had by then performed a valuable
service for the concrete block industry. Although
business activity in the 1930s was in a constantly
deepening trough of despair, lifted only by public
building programs, the decade was surprisingly
productive in a good many technological areas for the
concrete block industry.
C7.2.5 Mortar
The common variety of mortar was made of lime, sand,
and water. Details of its preparation varied according to
regional customs and individual preferences, but most
of these details were well known throughout Europe
and America. Sand was added to lime for economy, to
prevent shrinkage, and in such quantity that the lime
would fill the interstices. If an excess of sand was used,
the bond was poor. If too little sand was used, the
mortar would shrink and crack.
In ordinary sands, the spaces were from 39% to 40% of
the total volume, and in such, 1.0 volume of
cementitious paste (cement plus lime) would fill voids
of 2.5 volumes of sand. In practice, 1.25 to 2.0 volumes
of sand to 1.0 of paste was used. Thus, “pure” lime
mortar meant three to five volumes of sand to one
measured volume of lime. This gave a plastic mortar
that did not crack.
Until about 1890, the standard mortar used for masonry
in the United States was a mixture of sand and pure lime
(i.e., hydraulic lime) or lime-pozzolon-sand.
Massachusetts Hall (1730) at Harvard University and
Independence Hall (1734) in Philadelphia were built
with lime mortars that were also known as “fact”
mortars. These low-strength mortars gave masonry a
low modulus of elasticity and, therefore, an ability to
absorb considerable strain without inducing high stress.
Accordingly, the tendency to crack was reduced, and
when cracks did appear, masonry of high lime-content
mortar was to a great extent capable of chemical
reconstitution, i.e., “autogenous healing.”
After 1819, all masonry used in the construction of the
Erie Canal was laid in natural cement mortar. Various
FEMA 274 Seismic Rehabilitation Commentary
sources afford different information about the mortar
mix; apparently one part of sand was mixed with two
parts of cement. The general practice in New York state
in about 1840 was to mix two or three parts of sand to
one of cement.
For natural cements, the proportion of sand to cement
by measurement usually did not exceed three to one,
and for piers and first-class work a ratio of two to one
was used. Portland cement mortar commonly contained
four parts of sand to one of cement for ordinary mortar,
and three to one for first-class mortar. For work under
water, not more than two parts of sand to one of cement
were used. When cheaper mortars than these were
desired, it was considered better to add lime to the
mortar than more sand. Cement mortars were
introduced about 1880. Joints of cement mortar were
strong and unyielding because of the cement; they were
appropriate for bonding to modern bricks and concrete
blocks.
C7.2.6 Reinforced Masonry
Reinforced brick masonry was first used by Marc
Isambard Brunel in 1825, in the building of the Thames
Tunnel in England (Plummer and Blume, 1953).
Reinforced brick masonry was used by many builders
during that century; however, these builders were
individuals who had a feel for materials and built their
structures based upon their experience, more as an art
than from a rational design. Prior to 1880, a few
attempts were made to develop design formulas.
However, the performance of composite steel and
masonry flexural members was not clearly understood
at that time, and many investigators have attributed the
strength of the construction primarily to the adhesive
properties of the masonry. In fact, most of the early tests
were designed to demonstrate the increased strength
obtainable through the use of a new Portland cement in
mortar, instead of the hydraulic limes and natural
cements formerly used.
In the United States, Hugo Filippi, C.E. built and tested
reinforced brick masonry beams in 1913. Later in 1919,
L.J. Mensch, C.E. of Chicago also tested reinforced
brick beams in which the reinforcement was placed in a
bed of mortar below the brick masonry. However, the
data from these tests and others were never published
and there was little, if any, exchange of information
among those interested in the subject.
In 1923, the Public Works Department of the
Government of India published Technical Paper #38, a
7-3
 Chapter 7: Masonry (Systematic Rehabilitation)
comprehensive report of extensive tests of reinforced
brick masonry structures extending over a period of
about two years. A total of 282 specimens were tested,
including reinforced brick masonry slabs of various
thicknesses, reinforced brick beams, both reinforced
and unreinforced columns, and reinforced brick arches.
These tests appeared to be the first organized research
on reinforced brick masonry; the data provided answers
to many questions regarding this type of construction.
This research may therefore be considered as marking
the initial stage of the modern development of
reinforced brick masonry.
The idea of using cement-sand grout instead of bonding
brick headers to bind brick wythes or tiers together, and
inserting reinforcing steel in the grout space for tensile
and shearing resistance, was developed for practical and
sound engineering use in southern California beginning
about 1935. Since then, thousands of tests have been
conducted on full-size beams, slabs, and walls, from
which sound engineering design criteria have been
established and incorporated into building codes
throughout the United States.
C7.3 Material Properties and
Condition Assessment
C7.3.1 General
The term “masonry” is used to define the composite of
units, mortar, and possibly grout and/or reinforcement.
Whereas there are specifications to control the
manufacture of each of the constituent materials, the
most basic engineering properties to consider for
analysis of a building system are those representing the
composite. Thus, permissible values are given in this
section for compressive strength and elastic modulus of
the masonry assemblage, flexural tensile strength at the
unit-mortar interface, and shear strength and shear
modulus of vertical components such as piers, panels,
and walls. These mechanical properties will be relied on
for estimating stiffness and strength of masonry wall
and infill components.
C7.3.2 Properties of In-Place Materials
C7.3.2.1 Masonry Compressive Strength
Three options are given for measuring expected
masonry compressive strength. The first two methods
rely on testing of either extracted or rebuilt masonry
prisms in a laboratory. The third method measures
7-4 Seismic Rehabilitation Commentary
strength in situ by inserting a pair of flat jacks in an
existing masonry wall.
For the first method, sample test prisms are extracted
from a masonry component and transported to a
laboratory. The test prisms are subjected to vertical
compressive stress until the peak strength is reached.
The prism height should be at least twice its thickness,
contain at least two bed joints, and be a minimum of
15 inches high. The advantage of this method is that an
actual prism can be tested under controlled laboratory
conditions. In addition, strains can be monitored to infer
the elastic modulus (see Section C7.3.2.2). The
disadvantages are that the compressive strength might
be reduced during extraction, and the number of test
specimens is limited because of the cost of both the
extraction and the repair of the wall.
The second method requires test prisms to be fabricated
from actual masonry units that are extracted from an
existing masonry component. A chemical analysis of
the mortar is required so that mix proportions can be
simulated, and the mortar can be recreated. The
advantage of this method is the same as for the first
method. The disadvantage is that long-term creep,
moisture, and temperature effects cannot be simulated.
The third method consists of cutting slots in two mortar
bed joints, four to six courses apart, so that thin,
hydraulic flat jacks can be inserted and pressurized. The
portion of the masonry between the jacks is subjected to
a state of vertical compressive stress. The jacks are
stressed until the strength of the masonry is reached.
For masonry that is relatively weak, softening can be
observed by a reduction in slope of the stress-strain
curve, and compressive strength can be inferred. The
advantage of this method is that it is nondestructive and
the strength is measured in situ. In addition, the test can
be done in concert with other tests done to measure
elastic modulus and in situ compressive stress. The
disadvantage is that typical flat jacks may not be able to
reach the high pressures needed to approach the
ultimate strength of the masonry in compression.
As an alternative to the test methods given in the
Guidelines, the expected masonry compressive strength
may be deduced from a nominal value prescribed by the
Masonry Standards Joint Committee specification for
new construction (MSJC, 1995) knowing the unit
strength and mortar type (Specification Table 1 for clay-
unit masonry and Table 2 for concrete masonry). Tests
of extracted masonry units may be done to ascertain the
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
unit strength, or conservative estimates of unit strength
can be assumed for use with the MSJC tables. Likewise,
mortar type can be evaluated experimentally or
assumed. The MSJC table values are based on data
from masonry constructed after the 1950s and are only
applicable to this period of construction. Many of the
earlier mortars were lime-based rather than cement-
based as assumed with these table values. Furthermore,
earlier mortars were classified with a different
nomenclature than given in these tables, making direct
relations difficult. Therefore, the unit-strength
procedure using the MSJC tables should only be used
for masonry constructed after 1960. Expected masonry
strength should be determined by multiplying Table 1
values by a factor of 2.0 or Table 2 values by a factor of
1.5. These approximate factors are based on estimated
ratios between expected and lower bound compressive
strengths, as well as on correction factors for clay brick
and concrete block prisms.
Default values of compressive strength are set at very
low stresses to reflect an absolute lower bound.
Masonry in poor condition is given a strength equal to
one-third that for masonry in good condition, to reflect
the influence of mortar deterioration and unit cracking
on compressive strength.
C7.3.2.2 Masonry Elastic Modulus in
Compression
The elastic modulus of masonry in compression can be
measured by one of two methods. Each method
measures vertical strain between two gage points to
infer strain, and thus elastic modulus. The first method
consists of extracting a test prism from an existing wall;
the second method utilizes a pair of flat jacks to subject
an in situ portion of masonry to vertical compressive
stress.
The extracted prism method is essentially the same as
for the compressive strength test, with the difference
that dial gages or electronic displacement transducers
are fixed on the test prism to measure strain between
two gage points.
The flat-jack method is done in the same way as for the
compressive strength test, with the difference that the
jacks are pressured to less than half the masonry
strength. Vertical contractions of the compression field
between the two jacks are measured with a mechanical
dial gauge or electronic displacement transducer. Strain
is then determined by dividing measured distortion by
the length between gauges. Using correction factors for
FEMA 274 Seismic Rehabilitation Commentary
shape and stiffness of flat jacks, vertical compressive
stress is inferred from measured hydraulic pressure. The
elastic modulus, Eme, is calculated as the slope of the
stress-strain curve between 5% and 33% of the
estimated masonry ultimate compressive strength.
The flat-jack method has been shown to be accurate
within 10%, based on correlations between test values
and measured elastic moduli of test prisms (Epperson
and Abrams, 1989; Noland et al., 1987). A case study
using this method is presented by Kariotis and Ngheim
(1995). An available standard is the Standard Test
Method for In-Situ Elastic Modulus within Solid Unit
Masonry Estimated Using Flat Jack Measurements,
ASTM C 1197.
Default values of elastic modulus shall be based on a
coefficient of 550 times the expected masonry
compressive strength. This coefficient is set lower than
previous values given in the Uniform Building Code to
compensate for larger values of expected strength.
C7.3.2.3 Masonry Flexural Tensile Strength
Although the flexural tensile strength of older brick
masonry walls constructed with lime mortars may often
be neglected, the tensile strength of newer concrete and
clay-unit masonry walls can result in appreciable
flexural strengths. Therefore, guidelines for measuring
flexural tensile strength in situ or from extracted
specimens are given in this section.
Masonry flexural tensile strength can be measured
using a device known as a bond wrench, which clamps
onto the top course of a test specimen and applies a
weak-axis bending moment until the top masonry unit
snaps off. Flexural tensile stress is inferred by dividing
the moment capacity by the section modulus of the wall
section. The test can be done on test specimens
extracted from an existing wall, or in situ on a portion
of masonry that has been isolated by cutting vertical
slots on either side of the test portion. Alternatively,
flexural tension stress can be measured by bending
extracted portions of a masonry wall across a simply
supported span.
For the field test, two adjacent units of a running bond
pattern are removed so that a clamp may be inserted.
Single masonry units above and below the removed
units are subjected to an out-of-plane moment using a
calibrated torque wrench. Mortar head joints on either
sides of the tested units are removed to isolate the test
7-5
 Chapter 7: Masonry (Systematic Rehabilitation)
units. The laboratory test is done in much the same
manner on specimens that are cut from a wall. Test
prisms should be at least two units in height, and one
unit long, or a minimum of four inches. Both methods
involve substantial repair of the existing wall. An
available standard for the laboratory method is
Standard Test Methods for Masonry Bond Wrench
Testing, ASTM C 1072. No standards exist on the field
bond wrench test; however, this ASTM standard should
suffice.
The third method consists of extracting sample panels
or prisms from an existing masonry wall, and subjecting
them to minor-axis bending with either a third-point
loading or a uniform load distribution with an airbag.
Flexural tensile strength is determined by dividing the
maximum applied moment by the section modulus of
the masonry section. Standard Test Methods for
Masonry Flexural Tension Stress, ASTM E 518, is
available; however, ASTM does not recommend this
method for determination of design stresses.
For all three of these methods, the bonding of the test
unit to the mortar is sensitive to any disturbances that
are incurred during specimen removal. The confidence
level can be low because the scatter of data for flexural
bond strength can be high, and the number of test
samples is limited because of cost and the disturbance
concerns.
These test methods are intended for out-of-plane
strength of unreinforced masonry walls. For in-plane
bending, flexural stress gradients across the section
width are much lower than for out-of-plane bending.
Thus, data from tests described in this section should
not ideally be used for in-plane bending. However, in
lieu of data on in-plane tensile strength, out-of-plane
strength values may be substituted.
Default values for flexural tensile strength are set low
even for masonry in good condition, because of the
dependence of the unit-mortar bonding on the tensile
strength. This bonding can be highly variable,
depending on the relative absorption of the unit and the
water retentivity of the mortar, the presence and type of
cement used in the mortar, the previous loading history,
and the condition of the mortar. For masonry in poor
condition, a zero value of tensile strength is prescribed.
C7.3.2.4 Masonry Shear Strength
Expected shear strength of URM components can be
inferred from in situ measurements of bed-joint shear
7-6 Seismic Rehabilitation Commentary
strength using the in-place shear test. The
nondestructive test measures the in situ shear strength
between a clay masonry unit and the mortar bed joints
above and below the unit. A small hydraulic jack is
placed in a void left by removal of a masonry unit
immediately adjacent to the test unit. The head joint on
the opposite face of the test unit is removed to isolate
the test unit so that it may be displaced horizontally
when pushed.
A horizontal force is applied to the test unit until it starts
to slide. Shear strength is then inferred as the measured
force divided by the area of the bed joints above and
below the masonry unit. The estimated vertical
compressive stress at the test location is subtracted from
this value to give the bed joint shear stress, vto
(Equation 7-2), assuming a coefficient of friction equal
to 1.0. Because expected values of wall shear strength
are to be used, the 50th percentile value, vt , is used as
the index value.
The method is limited to tests of the face wythe. When
the test unit is pushed, resistance is provided across not
only the bed-joint shear planes, but also the collar-joint
shear plane. Because seismic shear is not transferred
across the collar joint in a multiwythe masonry wall, the
estimated shear resistance of the collar joint must be
deducted from the test values. This is done by
multiplying the vte term by 0.75 in Equation 7-1, which
is the ratio of the areas of the top and bottom bed joints
to the sum of the areas of the bed and collar joints for a
typical clay unit. If it is known that the collar joint is not
present, or is in very poor condition, the 0.75 factor may
be waived.
The effect of friction at the particular location of the
masonry element being evaluated is included by
increasing the bed-joint shear capacity by the addition
of the term “P/A” in Equation 7-1. The sum is then
multiplied by a reduction factor equal to 0.75, and
divided by 1.5 to convert it to an average stress for use
with walls of a rectangular cross section.
The in-place shear test was developed solely for solid
clay-unit masonry. However, the test method has been
used for single-wythe hollow concrete block masonry.
As for the conventional method with brick masonry, a
single unit is removed adjacent to a test unit as well as
the opposite mortar head joint. The maximum
horizontal force needed to move the block is divided by
the total area of the bed joint mortar above and below
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
the test unit and the total grouted area. The term vto is
obtained by subtracting the apparent vertical
compressive stress from this ratio as given in
Equation 7-2. If the shear capacity of the masonry
exceeds that of the loading equipment, the test may be
run on one-half the length of a block. In such case, the
mortar bed joints along one-half the length of the block
are removed.
An alternate in-place shear test method is to
simultaneously apply a vertical compressive stress,
using hydraulic flat jacks placed in the bed joints above
and below the test brick, while shearing the test brick.
In-place shear tests are done at various levels of vertical
compressive stress so that values of cohesion and
frictional coefficients can be inferred.
The available standard In-Place Masonry Shear Tests
(UBC Standard 21-6), is referenced in the 1994
Uniform Code for Building Conservation (ICBO,
1994), Appendix Chapter 1, Sections A106(c)3 and
A107(b).
Default values for shear strength of URM are provided,
ranging from 27 psi for good condition to 13 psi for
poor condition. If in-place shear tests are done, the
upper bound of vme by Equation 7-1 is 37 psi for a zero
vertical compressive stress when the 100 psi limit on vte
is considered. Thus, a 37% increase in strength is
possible if testing is done and the masonry is considered
to be in good condition. Default values for shear
strength of poor masonry are large relative to values for
masonry in good condition (1:2), because frictional
shear can be developed even when mortar or units are
deteriorated.
Shear strength of reinforced masonry (RM) cannot be
expressed in terms of the bed-joint shear stress because
of the influence of the vertical and horizontal
reinforcement on shear strength. There are no in situ
methods for measuring shear strength of existing RM
walls. Equations given for shear strength in BSSC
(1995) must be relied on. Ideally, the theory of
mechanics of materials does not change with age, and
the same strength equations should apply for existing or
new construction. However, care should be taken to
ensure that the condition of the existing masonry
components is comparable to that of newly constructed
elements. This assessment should include a review of
reinforcing details as well as the general condition of
the masonry (see Section 7.3.3).
FEMA 274 Seismic Rehabilitation Commentary
C7.3.2.5 Masonry Shear Modulus
Laboratory tests of URM shear walls (Epperson and
Abrams, 1989; Abrams and Shah, 1992) have found
that the shear modulus of masonry does approach the
value of 0.4 times the elastic modulus in compression,
as given by the theory of elasticity for isotropic, elastic
members. This value is limited to elastic, uncracked
behavior of the masonry. After cracking, the shear
stiffness is known to reduce substantially as sliding
along bed joints develops or as diagonal tension cracks
open. Because these nonlinear effects cannot be related
to the elastic modulus in compression, the 0.4Em value
is only appropriate for uncracked masonry. Shear
stiffness of post-cracked masonry can be taken as a
fraction of the initial shear stiffness. Test data by
Atkinson et al. (1989) provide estimates of shear
stiffness based on a frictional mechanism along bed
joints.
C7.3.2.6 Strength and Modulus of
Reinforcing Steel
The expected strength of reinforcing bars can be best
determined from tension tests of samples taken from the
building. If available, mill test data for the reinforcing
steel used in the building may be substituted.
Default values of yield strength are given to be the same
as for reinforcing bars in reinforced concrete (see
Section 6.3.2.5).
C7.3.2.7 Location and Minimum Number of
Tests
The required number of tests have been established
based on theories of statistical sampling, and past
experience.
C7.3.3 Condition Assessment
The goals of a condition assessment are:
• To examine the physical condition of primary and
secondary components and the presence of any
degradation
• To verify the presence and configuration of
components and their connections, and continuity of
load paths between components, elements, and
systems
• To review other conditions, such as neighboring
party walls and buildings, presence of nonstructural
7-7
 Chapter 7: Masonry (Systematic Rehabilitation)
components, and limitations for rehabilitation, that
may influence building performance
• To formulate a basis for selecting a knowledge
factor
The physical condition of existing components and
elements, and their connections, should be examined for
deterioration of masonry units, mortars, grouts, and
reinforcement. Deterioration may include
environmental effects (e.g., fire damage, chemical
attack, freeze/thaw damage) or past/current loading
effects (e.g., overload, damage from past earthquakes,
cracking). Masonry construction is also susceptible to
expansion and contraction due to thermal and moisture
conditions.
A condition assessment should examine configuration
problems such as discontinuous reinforcement patterns,
unequal alignment of components, and inadequate
connections between walls and foundation.
The scope of a condition assessment shall include an
investigation of primary and secondary structural
elements and components. Although masonry veneer is
not part of the structural system, the condition and
attachment of the veneer should be examined.
Substantial damage to masonry veneer has been
observed in numerous earthquakes (Klingner, ed.,
1994). Rehabilitation measures should be undertaken to
mitigate damage to veneer. However, since the veneer
is not part of the structural system, such measures will
not involve the Systematic Rehabilitation procedures
prescribed in Chapter 7. Accessibility constraints may
necessitate the use of instruments such as a fiberscope
or video probe, to reduce the amount of damage to
covering materials and fabrics. The knowledge and
insight gained from the condition assessment are
invaluable to the understanding of load paths and the
ability of components to resist and transfer these loads.
Destructive or nondestructive test methods may be
necessary to examine the interior portions of a masonry
structural component. Local removal of sheathing or
coatings on masonry wall surfaces may need to be done
to expose connections between the masonry and
adjoining components. The number of such
examinations will vary with the complexity and
availability of construction drawings.
7-8 Seismic Rehabilitation Commentary
C7.3.3.1 Visual Examination
Visual observations are simple and generally
inexpensive, and can detect obvious condition states in
the masonry materials and quality of construction.
Configuration problems can quickly be identified with
direct visual inspection. The continuity of load paths
can be determined through viewing of components and
connection condition. Visual inspection can determine
the need for other test methods to quantify the presence
and degree of deterioration.
The process of establishing component properties
should start with obtaining construction documents.
Preliminary review of these documents should be done
to identify primary gravity- and lateral-load-carrying
elements, systems, components, and connections. In the
absence of a complete set of building drawings, a
thorough inspection of the building should be done to
identify all load-bearing systems, and an as-built set of
drawings should be made.
If coverings or other obstructions exist, indirect visual
inspection can be done through use of drilled holes and
a fiberscope.
C7.3.3.2 Nondestructive Tests
Four tests are recommended to assess the relative
condition of masonry components: ultrasonic pulse
velocity, mechanical pulse velocity, impact echo, and
radiography. Merits and limitations of each method are
described in this section. Further information can be
found in Abrams and Matthys (1991).
A. Ultrasonic Pulse Velocity
Measurement of the velocity of ultrasonic pulses
through a wall can detect variations in the density and
modulus of masonry materials as well as the presence of
cracks and discontinuities. Transmission times for
pulses traveling through a wall (direct method) or
between two points on the same side of a wall (indirect
method) are measured and used to infer wave velocity.
Test equipment with wave frequencies in the range of
50 kHz has been shown to be appropriate for masonry
walls. Use of equipment with higher-frequency waves
is not recommended because the short wave length and
high attenuation are not consistent with typical
dimensions of masonry units.
Test locations should be sufficiently close to identify
zones with different properties. Contour maps of direct
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
transmission wave velocities can be constructed to
assess the overall homogeneity of a wall elevation. For
indirect test data, vertical or horizontal distance can be
plotted versus travel time to identify changes in wave
velocity (slope of the curve). Abrupt changes in slope
will identify locations of cracks or flaws.
Ultrasonic methods are not applicable for masonry of
poor quality or low modulus, or with many flaws and
cracks. The method is sensitive to surface condition, the
coupling material used between the transducer or
receiver and the brick, and the pressure applied to the
transducer.
The use of ultrasonic pulse velocity methods with
masonry walls has been researched extensively (Calvi,
1988; Epperson and Abrams, 1989; Kingsley et al.,
1987). A standard for the use of ultrasonic methods for
masonry is currently under development in Europe with
RILEM Committee 76LUM.
B. Mechanical Pulse Velocity
The mechanical pulse velocity test consists of
impacting a wall with a hammer blow and measuring
the travel time of a sonic wave across a specified gage
distance. An impact hammer is equipped with a load
cell or accelerometer to detect the time of impact. A
distant accelerometer is fixed to a wall to detect the
arrival time of the pulse. Wave velocity is determined
by dividing the gage length by the travel time. The form
and duration of the generated wave can be varied by
changing the material on the hammer cap.
The generated pulse has a lower frequency and higher
energy content than an ultrasonic pulse, resulting in
longer travel distances, and less sensitivity to small
variations in masonry properties and minor cracking.
The mechanical pulse method should be used in lieu of
the ultrasonic pulse method when overall mean
properties of a large portion of masonry are of interest.
The use of mechanical pulse velocity measurements for
masonry condition assessments has been confirmed
through research (Epperson and Abrams, 1989;
Kingsley et al., 1987). Although no standard exists for
mechanical pulse velocity tests with masonry, a
standard for concrete materials does exist, which may
be referenced: Test Method for Pulse Velocity through
Concrete (10-150 kHz range), ASTM C 597.
FEMA 274 Seismic Rehabilitation Commentary
C. Impact Echo
The impact-echo technique can be useful for
nondestructive determination of the location of void
areas within grouted reinforced walls (Sansalone and
Carino, 1988). Commercial devices are available or
systems can be assembled using available electronic
components. Since this technique cannot distinguish
between a shrinkage crack at the grout-unit interface
and a complete void in the grout, drilling of small holes
in the bed joint or examination using an optical
borescope should be performed to verify the exact
condition.
D. Radiography
A number of commercial devices exist that can be used
to identify the location of reinforcing steel in masonry
walls. They are also useful for locating bed-joint
reinforcing steel, masonry ties and anchors, and
conduits and pipes. The better devices can locate a No.
6 bar at depths up to approximately six inches; however,
this means that for a 12-inch-thick concrete masonry
wall, a bar located off-center cannot be found when
access is limited to only one side of the wall. These
devices are not able to locate or determine the length of
reinforcing bar splices in walls for most cases. They
work best for identifying the location of single isolated
bars, and become less useful when congestion of
reinforcing bars increases.
C7.3.3.3 Supplemental Tests
A. Surface Hardness
The surface hardness of exterior-wythe masonry can be
evaluated using the Schmidt rebound hammer. Research
has shown that the technique is sensitive to differences
in masonry strength, but cannot by itself be used to
determine absolute strength. A Type N hammer
(5000 lb.) is recommended for normal-strength
masonry, while a Type L hammer (1600 lb.) is
recommended for lower-strength masonry. Impacts at
the same test location should be continued until
consistent readings are obtained, because surface
roughness can affect initial readings.
The method is limited to tests of only the surface wythe.
Tuckpointing may influence readings and the method is
not sensitive to cracks.
Measurement of surface hardness for masonry walls has
been studied (Noland et al., 1987).
7-9
 Chapter 7: Masonry (Systematic Rehabilitation)
B. Vertical Compressive Stress
In situ vertical compressive stress resisted by the
masonry can be measured using a thin hydraulic flat
jack that is inserted into a removed mortar bed joint.
Pressure in the flat jack is increased until distortions in
the brickwork are reduced to the pre-cut condition.
Existing vertical compressive stress is inferred from the
jack hydraulic pressure, using correction factors for the
shape and stiffness of the flat jack.
The method is useful for measurement of gravity load
distribution, flexural stresses in out-of-plane walls, and
stresses in masonry veneer walls that are compressed by
a surrounding concrete frame. The test is limited to only
the face wythe of masonry.
Not less than three tests should be done for each section
of the building for which it is desired to measure in situ
vertical stress. The number and location of tests should
be determined based on the building configuration, and
the likelihood of overstress conditions.
C. Diagonal Compression Test
A square panel of masonry is subjected to a
compressive force applied at two opposite corners along
a diagonal until the panel cracks. Shear strength is
inferred from the measured diagonal compressive force
based on a theoretical distribution of shear and normal
stress for a homogeneous and elastic continuum. Using
the same theory, shear modulus is inferred from
measured diagonal compressive stress and strain.
Extrapolation of the test data to actual masonry walls is
difficult because the ratio of shear to normal stress is
fixed at a constant ratio of 1.0 for the test specimens.
Also, the distribution of shear and normal stresses
across a bed joint may not be as uniform for a test
specimen as for an actual wall. Lastly, any
redistribution of stresses after the first cracking will not
be represented with the theoretical stress distributions.
Thus, the test data cannot be useful to predict nonlinear
behavior.
If the size of the masonry units relative to the panel
dimension is large, masonry properties will be not
continuous, but discrete. Test panels should be a
minimum of four feet square. The high cost and
disruption of extracting a number of panels this size
may be impractical.
A standard is available, titled Standard Test Method for
Masonry Diagonal Compression, ASTM E 519.
7-10 Seismic Rehabilitation Commentary
D. Large-Scale Load Tests
Large-scale destructive tests may be done on portions of
a masonry component or element to (1) increase the
confidence level on overall structural properties, (2)
obtain performance data on archaic building materials
and construction materials, (3) quantify effects of
complex edge and boundary conditions around
openings and two-way spanning, and (4) verify or
calibrate analytical models. Large-scale load tests do
not necessarily have to be run to the ultimate limit state.
They may have value for simply demonstrating
structural integrity up to some specific Performance
Level.
Out-of-plane strength and behavior of masonry walls
can be determined with air-bag tests. Behavior of test
panels incorporating connections and edge details can
be determined from such a test, in addition to flexural
and arching properties of a solid or perforated wall.
Strength and deformation capacity under in-plane
lateral forces can be determined by loading an
individual portion of wall that is cut free of the
surrounding masonry. Loading actuators are reacted
against adjacent and stronger portions of masonry. Such
testing is particularly useful when the wall is composed
of different materials that cannot be evaluated by testing
an individual unit of an individual wythe.
Visual and nondestructive surveys should be used to
identify locations for test samples.
Standards for laboratory test methods are published by
ASTM. Procedures for removal and transportation of
masonry samples are given in Evaluation of Structural
Properties of Masonry in Existing Buildings, NBS
Building Science Series 62, U.S. Department of
Commerce.
Large-scale tests are expensive and limited to a single
or few samples. They may result in considerable local
damage and may require substantial reconstruction near
the sample location. Test data must be extrapolated to
the remainder of the system based on a low confidence
level.
C7.3.4 Knowledge (κ ) Factor
The level of knowledge of a particular masonry
structure may conform to either a minimum level or an
enhanced comprehensive level. As noted in
Section 2.7.2, knowledge factors, κ, are assigned equal
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
to 0.75 and 1.00 for these two levels. The Linear Static
Procedure (LSP) of Chapter 3 may be used with either
knowledge level, but the Nonlinear Static Procedure
(NSP) is limited to a κ factor equal to 1.0.
The basic distinction between the two levels of
knowledge is whether or not in situ tests of masonry
materials are done. For the minimum level, a visual
examination of the structure is required per
Section 7.3.3.1; however, in-place testing is not
necessary. Thus, the LSP may be used with the default
values of material strengths as specified in
Section 7.3.2. For the comprehensive level of
knowledge, some in situ material testing is required in
addition to the nondestructive testing for condition
assessment noted in Section 7.3.3.2. These tests include
determination of masonry compressive strengths using
one of the methods prescribed in Section 7.3.2.1 for
both unreinforced and reinforced masonry. For
unreinforced masonry only, in-place shear strength tests
must be done in accordance with Section 7.3.2.4. For
reinforced masonry only, tensile strengths of reinforcing
bars must be determined in accordance with
Section 7.3.2.6.
Even for the comprehensive level of knowledge, in situ
tests of masonry flexural tensile strength or elastic
modulus are not required. This is because tensile
strength should be quite low and somewhat similar to
the default values as given in Section 7.3.2.3. Similarly,
test data for elastic modulus can have a large scatter and
not differ from the approximate value given in
Section 7.3.2.2 (550 times the masonry expected
compressive strength).
C7.4 Engineering Properties of
Masonry Walls
Masonry building systems are composed largely of
walls. Masonry walls may be divided between
structural walls—such as bearing or shear walls—and
nonstructural walls, such as partition walls, cladding,
veneer, infills, and parapets. Engineering properties
given in Section 7.4 apply only to structural walls.
Masonry bearing walls support floor and roof gravity
loads, and may or may not be shear walls. Conversely,
masonry shear walls resist lateral seismic forces, and
may or may not be bearing walls. If a wall is part of the
lateral-force-resisting system, it is considered as a
primary element. If the wall supports only gravity loads
FEMA 274 Seismic Rehabilitation Commentary
and must remain stable under lateral sway, it is
considered as a secondary element. All other masonry
walls are excluded from Section 7.4.
C7.4.1 Types of Masonry Walls
Structural masonry walls are classified into three
fundamental types: existing, new, and enhanced.
Guidelines for determining structural properties of
masonry walls reference current standards, which are
different for existing and new walls. In addition, the
Guidelines provide specific recommendations on
minimum requirements for enhancement of existing
walls so that their structural properties may be
considered the same as those of new or existing
elements or components.
Rehabilitated buildings typically consist of lateral-
force-resisting systems that comprise a combination of
different materials. An existing unreinforced masonry
building might be strengthened by adding braced steel
frames, or conversely, a new reinforced masonry wall
might be added to stiffen a flexible steel frame. Existing
masonry walls might be enhanced with shotcrete or
surface coatings, reinforced or prestressed cores, grout
injections, or repointing, or by changing the size of
openings. The engineering properties given in
Section 7.4 are applicable to building systems with
existing, new, or enhanced masonry walls that combine
to rehabilitate a building system.
Stiffness assumptions, strength criteria, and acceptable
deflections for various limit states as described in
Sections 7.4.2 through 7.4.5 are common for existing,
new, or enhanced masonry walls. Principles of
mechanics are the same despite the age of a masonry
wall. Physically, there should be no difference in
stiffness assumptions, strength criteria, or inelastic
behavior for existing, newly constructed, or enhanced
walls. Thus, guidelines on determining engineering
properties for each of the three fundamental wall types
are expressed in common in these sections.
In Sections 7.4.2 through 7.4.5, walls are grouped in
terms of how they respond to lateral forces.
Unreinforced walls are presented first, followed by
reinforced walls, because the behavior of each type of
wall is distinctively different. Furthermore, walls
subjected to in-plane lateral forces are separated from
walls subjected to out-of-plane forces because their
stiffnesses, strengths, and acceptable deformations vary
widely.
7-11
 Chapter 7: Masonry (Systematic Rehabilitation)
C7.4.1.1 Existing Masonry Walls
Existing masonry walls will have a significant influence
on the lateral strength and drift of a building system.
Certain masonry walls may have a brittle character, and
partial or complete removal may improve the overall
energy dissipation capabilities of a system, and may
thus be a viable rehabilitation option. When considering
a particular rehabilitation scheme, existing masonry
walls, or their extraction, should be included in the
structural analysis along with any new masonry walls
that may be added.
A thorough condition assessment of existing masonry
walls should be made to increase the level of confidence
in characterizing structural properties.
C7.4.1.2 New Masonry Walls
Newly constructed masonry walls can be added to an
existing building system for the purpose of
strengthening, stiffening, or increasing inelastic
deformation and energy dissipation capacity. The
design of new masonry walls must follow the NEHRP
Recommended Provisions (BSSC, 1995). This standard
is based on strength design for both unreinforced and
reinforced masonry walls. When used in combination
with existing walls, no capacity reduction, or φ factors,
should be used.
In zones of high seismicity, new masonry walls must be
reinforced with at least the minimum percentages of
reinforcement as specified for a reinforced wall in
Section 7.8 (BSSC, 1995). In zones of moderate
seismicity, masonry walls must have a minimum of trim
bars at corners, top and bottom and around all openings
per the NEHRP Recommended Provisions.
Unreinforced walls can be added to an existing building
in zones of low seismicity since they are recognized by
this standard.
C7.4.1.3 Enhanced Masonry Walls
Both reinforced and unreinforced walls may be
rehabilitated by the various means noted in this section
to increase their strength, stiffness, and/or deformation
resistance capacity. Enhancement methods are not listed
in a priority order, nor are they necessarily the sole
methods that can be used.
A. Infilled Openings
A common method of stiffening or strengthening an in-
plane masonry wall is to fill window or door openings
7-12 Seismic Rehabilitation Commentary
with masonry. This is typically done for unreinforced
walls, but may also be applicable to reinforced walls if
needed.
Infilling of an existing opening will stiffen and
strengthen a perforated shear wall. The restriction of
opening length to no more than 40% of the overall wall
length was intended to limit the introduction of new
masonry, which by this provision may be considered to
exhibit behavior equal to that of the original masonry.
The percentage was chosen so that the majority of
masonry would be original.
B. Enlarged Openings
Door and window openings in unreinforced masonry
walls may be enlarged to alter the aspect ratio of an
adjacent pier. By removing a portion of masonry above
or below an opening, the height-to-length aspect ratio of
the adjacent piers will be increased to such an extent
that rocking behavior may govern their response.
Although this approach will weaken a perforated
masonry wall, it will also increase its inelastic
deformation capacity if a ductile rocking mechanism
can be invoked. Furthermore, if the method is used,
excessive diagonal tensile stresses can be relieved for a
relatively stocky pier, thus lowering its vulnerability to
nonductile “X” cracking.
The method is also applicable to infill panels.
Increasing the size of an opening will reduce infill
strength and stiffness and may relieve a surrounding
frame from excessive frame-infill interactive forces.
C. Shotcrete
Application of reinforced shotcrete to the surface of a
masonry wall is a common method for enhancing both
in-plane and out-of-plane strength. The shear area of the
wall is increased and the height-to-thickness (h/t) ratio
is lowered. Reinforcement embedded in the shotcrete
layer substantially improves both the shear and flexural
capacities. The method may be used with existing
reinforced masonry walls, but has its greatest potential
with unreinforced walls.
If shotcrete is used to enhance out-of-plane strength,
flexural behavior will be asymmetrical for loading in
each direction, since the compression zone will
alternate between the shotcrete layer and the masonry.
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
D. Coatings for URM Walls
Surface coatings may be used to enhance the in-plane
shear strength of a URM wall. The h/t ratio will be
reduced with the coating, which will enhance the
strength of the wall in compression and under
transverse loads. Coatings may consist of a cement
plaster coating with an embedded steel mesh, or a
gypsum plaster coating.
Research has been done on the effectiveness of using
fiber-reinforced composites (e.g., kevlar, carbon fibers)
for strengthening masonry walls; however, long-term
durability remains questionable.
E. Reinforced Cores for URM Walls
Existing URM walls may be reinforced in the vertical
direction by grouting reinforcing bars in cores drilled
through the wall height. The method, commonly known
as the “center core technique,” has been used
predominantly in California for seismic rehabilitation of
URM buildings. With adequate anchorage of new
vertical reinforcing bars in the drilled cores, a wall may
be assumed to act as a reinforced wall in flexure.
The use of epoxy resins to fill cores around reinforcing
bars in older, softer masonry materials has resulted in
accelerated deterioration due to incompatibility of
materials.
F. Prestressed Cores for URM Walls
Existing URM walls may be prestressed in the vertical
direction with strands or rods embedded at their base in
grout and placed in cores drilled through the wall
height.
Tendons should be ungrouted. Walls enhanced with
unbonded tendons will respond in a nonlinear but
elastic (returning to undeformed shape) manner. If
tendons are bonded with grout, inelastic straining of the
tendon can dissipate substantial seismic energy.
However, because of the high strength of most tendon
steel (cables or bars), excessive compressive strain may
result in premature crushing of the masonry before the
tendon can develop post-yield strains. Thus, hysteretic
damping and ductile performance will be inhibited.
Losses in prestressing force can be estimated based on
the expected shortening of a masonry component due to
elastic deformations, creep, and shrinkage effects.
Design procedures for estimating losses are given in
Curtin et al. (1988). Research results on creep and
FEMA 274 Seismic Rehabilitation Commentary
shrinkage movements of clay-unit masonry can be
found in Lenczner (1986).
Unlike the reinforced core technique, the prestressed
core technique will improve shear strength as well as
flexural strength because of the friction that is
developed as a result of the increased vertical
compressive stress.
G. Grout Injections
The shear strength of existing masonry walls can be
enhanced by injecting grout into the interior voids of the
wall. For unreinforced brick masonry walls, grout can
be injected into possible voids in the collar joint in
addition to the head and bed joints. This will also
increase the shear and tensile strength between wythes
and increase the transverse strength of a multiwythe
wall. For hollow-unit masonry, grout can be injected
into the open cells.
H. Repointing
Repointing is the process of removing deteriorated
mortar joints and replacing with new mortar.
Repointing can be used to enhance shear or flexural
strength of a URM wall.
I. Braced Masonry Walls
Steel bracing elements can be provided to reduce the
span of a masonry wall bending in the out-of-plane
direction.
J. Stiffening Elements
Additional structural members can be added to enhance
the out-of-plane flexural stiffness and strength of a
masonry wall. Such members may be placed in the
vertical and/or horizontal direction.
C7.4.2 URM In-Plane Walls and Piers
Walls resisting lateral forces parallel to their plane are
termed “in-plane walls.”
Solid walls deflect as vertical cantilevered flexural
elements from the foundation. Tall slender in-plane
walls (height larger than length) resist lateral forces
primarily with flexural mechanisms. Squat walls
(height less than length) resist lateral forces primarily
with shear mechanisms.
Perforated walls can be idealized as a system of piers
and spandrel beams. If beams are sufficiently stiff in
7-13
 Chapter 7: Masonry (Systematic Rehabilitation)
bending, piers can be assumed to be fully restrained
against rotation at their top and bottom. If openings in a
perforated wall are relatively large, the wall system will
deflect as a cantilevered shear element from the
foundation. Pier distortions in flexure and shear will
result in story drifts with little rotation of the floor level.
The provisions of Section 7.4.2 apply to both
cantilevered shear walls and individual pier elements
adjacent to window or door openings. The difference in
rotational boundary conditions at the top of either walls
or piers is accounted for with an α factor that increases
the lever arm of the vertical compressive force about the
toe for a pier type component.
C7.4.2.1 Stiffness
A. Linear Elastic Stiffness
Force-deflection behavior of unreinforced masonry
shear walls is linear-elastic before net flexural tension
stresses at the wall heel exceed tensile strengths, or
diagonal tension or bed-joint sliding shear stresses
exceed shear strengths.
Laboratory tests of solid shear walls have shown that
behavior can be depicted at low force levels using
conventional principles of mechanics for homogeneous
materials. In such cases, the lateral in-plane stiffness of
a solid cantilevered shear wall, κ, can be calculated
using Equation C7-1:
1 diagonal path through masonry units. The former action
k = ------------------------------------
- (C7-1)
3
h eff h eff
--------------- + --------------
3E m I g A v G m
where:
heff = Wall height
A = Shear area
v
Ig = Moment of inertia for the gross section
representing uncracked behavior
E = Masonry elastic modulus
m
G = Masonry shear modulus
m lateral deflection.
Correspondingly, the lateral in-plane stiffness of a pier
between openings with full restraint against rotation at
its top and bottom can be calculated using
Equation C7-2:
7-14 Seismic Rehabilitation Commentary
1
k = ---------------------------------------- (C7-2)
3
h eff h eff
------------------ + ------------- -
12E m I g A v G m
where the variables are the same as for Equation C7-1.
Analytical studies done by Tena-Colunga and Abrams
(1992) have shown that linear-elastic models can be
used to estimate measured dynamic response of an
unreinforced masonry building excited during the 1989
Loma Prieta earthquake.
B. Nonlinear Behavior of URM Walls
As the lateral force is increased on a wall or pier
component, flexural or shear cracking—or a
combination of both—will occur, resulting in
deflections that are nonlinear with respect to the applied
forces. Nonlinear behavior of URM walls has been
shown to be dependent on the length-to-height (L/h)
aspect ratio and the amount of vertical compressive
stress.
Behavior of relatively stocky walls (L/h greater than
1.5) is typically governed by diagonal tension or bed-
joint sliding, depending on the level of vertical
compression, masonry tensile strength, and bed-joint
sliding shear strength. For walls governed by diagonal
tension, cracks can develop in either a stair-step pattern
through mortar head and bed joints, or a straight
occurs when the mortar is weak relative to the units; the
latter occurs when the converse is true. The stair-
stepped pattern is better for inelastic deformation
capacity because vertical compressive stress normal to
the bed joints will result in the development of frictional
forces that will remain active at nearly any amount of
lateral deflection. Walls governed by a weaker bed-joint
sliding shear strength will deform with either a
concentrated deformation at one or a few bed joints, or
a distribution deformation across several bed joints,
depending on the ratio of the cohesion and the frictional
coefficient. The inelastic deformability of this sliding
type of deformation is also enhanced by frictional
forces that remain nearly constant despite the amount of
In walls with a moderate aspect ratio (L/h between 1.0
and 1.5), considerable strength increases have been
observed after flexural cracks form at the heel of a wall
as the resultant vertical compressive force migrates
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
towards the compressive toe. As the effective section
decreases with progressive cracking, the wall element
softens, gradually generating a nonlinear force-
deflection relation. If the shear capacity is not reached,
the ultimate limit state for such walls is toe crushing.
Flexural tension strength at the wall heel does not limit
lateral strength. Results from experiments by Epperson
and Abrams (1992) and Abrams and Shah (1992) have
revealed these tendencies. An analytical study by Xu
and Abrams (1992) investigated lateral strength and
deflection of cracked unreinforced masonry walls
behaving in this range.
For more slender walls (L/h less than 1.0) loaded with a
relatively light amount of vertical compressive force,
flexural cracks will develop along a bed joint near the
base of the wall. When the lateral force approaches a
value of PL/2h, the wall will start to rock about its toe,
provided that the shear strength will not be reached. A
singularity condition will exist momentarily as the
compressive stress at the wall toe increases rapidly just
before rocking, which will cause, at worst, some slight
cracking at the toe. Despite the fact that a bed-joint
crack will develop across almost all of the wall base, the
wall can still transfer shear because of friction at the
wall toe as a result of the vertical compressive force.
After rocking commences, the wall can be displaced to
very large drifts with no further damage as a result of
the rigid-body rotation about its toe. Again, flexural
tension strength at the wall heel does not limit lateral
strength. Behavior in this range has been observed with
experiments by Calvi et al. (1996) and Costley and
Abrams (1995).
The same types of action can be depicted for pier
components; however, the vertical compressive force
will shift towards the compression toe at both the top
and bottom of the pier. This restraining action will
cause the rocking strength to almost double because of
the increase in lever arm distance between the vertical
force couple. The use of the α factor in Equation 7-4,
which accounts for differences in rocking strengths for
cantilevered walls and fixed-fixed piers, is explained in
Kingsley (1995).
Upon unloading, wall or pier components subjected to
rocking actions will resume their original position as a
result of the restoring nature of the vertical compressive
force. For components subject to bed-joint sliding, the
slope of the unloading portion of the force-deflection
relation will be steep and will continue after the sense of
the deflection is reversed. Unlike a reinforced concrete
FEMA 274 Seismic Rehabilitation Commentary
or masonry beam, the hysteresis loop will not be
pinched. Thus, the area enclosed by the loop can be
large.
C. Lateral Stiffness with Linear Procedures
The linear procedures of Section 3.3 are based on
unreduced lateral forces for determination of
component actions. If the component is deformation-
controlled, these unreduced forces, QUD , are compared
with expected component strengths, QCE , multiplied by
m factors representing different ductilities. Because the
unreduced forces are fictitious, they cannot be used to
assess the expected amount of cracking in any
component. Thus, reductions in stiffness cannot be
estimated because actual force levels are not known.
Therefore, only initial, uncracked linear stiffnesses can
be used with the equivalent linear procedures. Any
nonlinear action is accounted for by applying the m
factor to expected strengths.
Much like that of a reinforced concrete beam past yield,
the tangent stiffness of a rocking wall or pier is quite
small relative to its uncracked stiffness before rocking.
For modeling the distribution of story shear to
individual piers, the linear stiffness is used rather than
the tangent rocking stiffness, which is analogous to the
procedure used for strength design of concrete
structures. Again, the initial stiffness is used to estimate
the elastic demand forces, which are then related to
expected strengths by introducing the m factor. Thus,
individual pier forces are not distributed in accordance
with rocking strengths—as is done with FEMA 178
(BSSC, 1992a) or UCBC procedures—but with respect
to relative elastic stiffnesses.
C7.4.2.2 Strength Acceptance Criteria
As noted in Section C7.4.2.1B, lateral strength of
unreinforced in-plane masonry walls or piers is limited
by diagonal tension, bed-joint sliding, toe crushing, or
rocking. Net flexural tension stress is not a limit for
strength, because post-cracked behavior is assumed for
the nonlinear range of response.
Rocking and bed-joint sliding are classified as
deformation-controlled actions because lateral
deflections of walls and piers can become quite large as
strengths remain close to constant. Diagonal tension
and toe crushing are classified as force-controlled
actions because they occur when a certain stress is
reached, and can cause sudden and substantial strength
deterioration. Stair-stepped diagonal cracking can also
7-15
 Chapter 7: Masonry (Systematic Rehabilitation)
be considered as a deformation-controlled action
because frictional forces along bed joints are conserved
with vertical compressive forces. However, diagonal
tension must be classified as a force-controlled action
unless stair-stepped cracking can be distinguished from
diagonal cracking through units.
A. Expected Lateral Strength of Walls and Piers
Expected bed-joint sliding shear strength is determined
using Equation 7-3. The expected bed-joint shear
strength from in-place shear tests is multiplied by the
full area of the mortar and/or grout. Although no shear
stress can be developed across flexural bed-joint cracks,
the increased compressive stress resisted by the
opposite wall or pier edge should compensate for this
reduction. For the case of a rocking pier, nearly all of
the bed joint may be open at the base and top to accept
the component’s rotation, yet shear is still transferred at
the toe because of friction.
Expected rocking strength of walls or piers is
determined using Equation 7-4, which was derived by
taking moments about the toe of the component. The
0.9 factor accounts for a slight reduction in the lever-
arm distance to represent the centroid of the vertical
compressive stress. If the component is a cantilevered
shear wall, the vertical axial compressive force is
assumed to act at the center of the wall at the top, which
is the reason for an α term equal to 0.5. If the
component is a pier, the vertical force is assumed to act
near its edge as the pier rotates and the superstructure
remains horizontal. The vertical compressive force,
PCE, should be the best estimate of the gravity force
during the earthquake.
Lateral strength of newly constructed masonry walls or
piers shall follow the NEHRP Recommended Provisions
(BSSC, 1995).
B. Lower Bound Lateral Strength of Walls and Piers
Lateral strength of walls or piers based on diagonal
tension strength is determined using Equation 7-5,
which is taken from Turnsek and Sheppard (1980). This
equation is only applicable for the range of L/h between
0.67 and 1.00. Because tests do not exist for masonry
diagonal tension strength, the bed-joint shear strength,
as measured with the in-place shear test, may be
substituted where it is assumed that the lower bound
diagonal tension strength is equal to the expected value
of the bed-joint strength.
7-16 Seismic Rehabilitation Commentary
Lateral strength limited by toe compression stress is
determined using Equation 7-6, which was derived
from Abrams (1992). The equation is only applicable
for walls or piers loaded with a lateral force that will not
result in rocking about their toe. It applies generally to
walls with L/h aspect ratios between 1.0 and 1.5 and
large vertical compressive stresses. For a lower bound
strength, a low estimate of vertical compressive force,
PCL, must be used. The limiting compressive stress is
conservatively taken as 93% of the lower bound
masonry compressive strength, f m ′ . Because the lower
bound strength is not determined per Section 7.3.2.1, it
may be estimated as a fraction of the expected
compressive strength, fme.
C. Lower Bound Vertical Compressive Strength of
Walls and Piers
The lower bound vertical compressive strength given by
Equation 7-7 includes a reduction factor equal to 0.85 to
relate prism strength to wall strength, and another factor
equal to 0.80 for accidental eccentricities.
C7.4.2.3 Deformation Acceptance Criteria
Unreinforced masonry walls or piers loaded parallel to
their plane may experience distress conditions of:
• Minor diagonal-tension or bed-joint cracking
• Major shear cracking or spalling of units
• Loss of strength
• Dislodgment and falling of units
• Out-of-plane movement as a result of excessive
rocking
The deformation acceptability criteria given in
Section 7.4.2.3 are intended to limit damage
accordingly for the goals of each Performance Level.
A. Linear Procedures
For the Linear Static Procedure, m factors are given for
primary and secondary components for each
performance level in Table 7-1.
As discussed in Section C7.4.2.1B., nonlinear force-
deflection behavior of unreinforced masonry shear
walls has been studied experimentally by a number of
researchers. Based on many of these wall tests, and
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
subjective but conservative interpretations of the test
data, the m factors given in Table 7-1 have been
derived. Because the experimental research is by no
means sufficiently complete to justify directly every
combination of wall aspect ratio and vertical
compressive stress, the m factors have been calibrated
in terms of an approximate value for a square wall panel
with a nominal amount of vertical compressive stress.
Therefore, for the Life Safety Performance Level, an m
value equal to 3.5 was established as a control point for
development of the table. This value is credible
considering that the test data revealed ductilities in
excess of five for wall panels with similar
characteristics.
Variable m factors are given for each Performance
Level, corresponding to approximate inelastic
deflections associated with specific damage states. For
Immediate Occupancy, some cracking can be tolerated
for typical occupancy conditions; m factors range from
1.0 for bed-joint sliding to 1.5 times the height-to-
length aspect ratio for a rocking mechanism. Larger
nonlinear displacements can be tolerated for rocking
piers because bed-joint cracks in rocking components
will close after an earthquake, whereas head-joint
cracks resulting from bed-joint sliding will not close
fully after the sliding stops. The height-to-length aspect
ratio is included in the m factor for rocking piers to
relate rigid-body rotation of a component to the lateral
deflection at the top of the component. The Life Safety
Performance Level is related to lateral deflections
associated with the dislodgment of masonry units and/
or severe cracking; m factors are conservatively set at a
value of 3.0 for bed-joint sliding or rocking of square
wall or pier components. The Collapse Prevention
Performance Level is related to a loss of lateral strength
for primary components, and unstable gravity-load
behavior of secondary components; therefore, m factors
are approximately one-third larger than for Life Safety.
B. Nonlinear Procedures
Nonlinear deformation capacities for primary and
secondary components are represented in Figure 7-1
with dimensions d and e respectively. These values are
consistent with the m values defined for each
Performance Level in Table 7-1, and have been
extracted from experimental studies on unreinforced
masonry walls as discussed in the previous section. The
wall drift before strength is lost (the d dimension in
Figure 7-1) is equal to 0.4% for bed-joint sliding or
rocking of square wall or pier components, which is
comparable to laboratory test values of approximately
FEMA 274 Seismic Rehabilitation Commentary
1% for walls that are governed by these deformation-
controlled actions. Drift levels have been reduced
substantially to 0.10% for walls with zero vertical
compressive stress because rocking or bed-joint sliding
mechanisms cannot be mobilized, and, as a result,
behavior will be governed by force-controlled actions
such as diagonal tension.
C7.4.3 URM Out-of-Plane Walls
Walls resisting lateral forces normal to their plane are
termed “out-of-plane walls.”
C7.4.3.1 Stiffness
Out-of-plane URM walls not subjected to significant
vertical compressive stress, and with no restraint at
boundaries for formation of arching mechanisms, do
not have a nonlinear range. They are brittle elements
that will crack under light lateral forces. Depending on
the particular Performance Level, cracking of a wall
panel may be acceptable if it can be shown that the wall
segments rotating about their ends will be stable under
dynamic loading.
The stiffness of walls bending about their weak axis is
three or more orders of magnitude less than the stiffness
of walls bending about their strong axis. Thus, in an
analysis of a building system with walls in each
direction, the stiffness of the transverse walls will be
much less than that of the in-plane walls and can
therefore be neglected.
C7.4.3.2 Strength Acceptance Criteria
Out-of-plane walls do not need to be analyzed using the
Linear Static Procedure because they act as isolated
elements spanning across individual stories. Rather than
design on the basis of an equivalent base shear applied
to the global structural system (per Equation 3-6 with
the Linear Static Procedure), out-of-plane walls should
resist inertial forces that are prescribed in Section 2.11.7
without cracking for the Immediate Occupancy
Performance Level. For similar reasons, the nonlinear
procedures are also not applicable for out-of-plane
walls.
The expected demand forces depend on response of the
floor or roof diaphragms and the in-plane walls. In
addition to the transverse inertial forces resulting from
the panel weight, a wall panel must also resist
deformations resulting from differential lateral drift
across a story, as well as diaphragm deflections. These
imposed deflections on the out-of-plane wall panels can
7-17
 Chapter 7: Masonry (Systematic Rehabilitation)
be accommodated with cracking of the bed-joints if
such cracking is determined to be acceptable for the
Performance Level. Even under small amounts of
vertical compressive stress, cracked panels will remain
stable as they deflect with the attached floor or roof
diaphragms.
The out-of-plane response of URM walls may be
governed by the development of arching mechanisms in
the vertical direction between the floor slabs above and
below, or in the horizontal direction between columns,
pilasters, or walls running in the normal direction. The
type of response mechanism for the out-of-plane wall
components is sensitive to the conditions at the panel
boundaries and the eccentricities of any applied vertical
loads. A rigorous analysis requires knowledge of:
• Accelerations of diaphragms above and below the
wall panel
• Edge restraint provided by slabs, beams, or
spandrels above and below the wall panel, and by
columns, pilasters, or walls at each side of the wall
panel
• Masonry compressive strength
• Mortar joint tensile strength
• Eccentricity of vertical compressive loads and
amounts of vertical load
In spite of these complexities, the out-of-plane strength
of URM walls may be bounded as follows.
• The lower limit of strength is defined for a wall
panel with no axial load other than its self weight, no
edge confinement from stiff elements above, below,
or to the sides, no continuity with adjacent wall
panels, and low tensile strength. If such conditions
are present, the out-of-plane static strength and
stiffness may be considered negligible. However, the
panel may be stable under dynamic action for the
Life Safety and Collapse Prevention Performance
Levels, as the weight of the panel tends to restore
lateral response back to its original position.
• The upper limit is defined for a wall panel that is
ideally fixed in one or two directions by walls,
columns, or pilasters that do not deflect, and vertical
compressive forces are applied concentrically about
the wall panel. Neglecting masonry tensile strength,
7-18 Seismic Rehabilitation Commentary
flexural cracking will commence when a uniform
transverse load, qcr , is applied equal to:
2Pt
q cr = --------- (C7-3)
2
h
where P is the vertical compressive load, and h and t
are the panel height and thickness. Because of
arching action, the panel can sustain transverse loads
with a reasonable upper bound of:
q cr = 6Pt
--------- (C7-4)
2
h
At the maximum load level, the wall stiffness can be
considered to be negligible; the structural integrity
of the panel is dependent on dynamic stability.
C7.4.3.3 Deformation Acceptance Criteria
Acceptance criteria for the Life Safety and Collapse
Prevention Performance Levels are based on stable
response after cracking of a wall panel has occurred. In
addition to the transverse inertial forces resulting from
the panel weight, a wall panel must also resist
deformations resulting from differential lateral drift
across a story, as well as diaphragm deflections. These
imposed deflections on the out-of-plane wall panels can
be accommodated with cracking of the bed-joints. Even
under small amounts of vertical compressive stress,
cracked panels will remain stable as they deflect with
the attached floor or roof diaphragms. Out-of-plane
response of cracked wall panels can be modeled
analytically with a dynamic analysis that implicitly
considers the motion input at the base of the wall and at
the top of the wall. Both the ground motion and the
motion of the diaphragm attached to the wall must be
determined for this analysis. Research (ABK, 1981) has
shown that wall segments should remain stable if their
h/t ratio is less than particular values. The values given
in Table 7-3, taken from Table C7.4.7.1 of BSSC
(1992), are quite conservative relative to the values
found in the ABK research. If the h/t ratio of an existing
wall exceeds the values given in Table 7-3, and a
dynamic stability analysis is not done, then the wall can
be either braced (see Section 7.4.1.3I) or thickened with
shotcrete (see Section 7.4.1.3C) or a surface coating
(see Section 7.4.1.3D). Conversely, the wall may be
reinforced (see Section 7.4.1.3E) and analyzed as a
reinforced wall, or the wall may be prestressed (see
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
Section 7.4.1.3F) to increase its cracking moment
capacity.
C7.4.4 Reinforced Masonry In-Plane Walls
and Piers
This section applies to reinforced wall and pier
components that resist lateral force parallel to their
plane. Information on modeling lateral stiffness and
expected strength of these components is given for
flexural, shear, and axial compressive actions.
As for unreinforced masonry wall and pier components
(Section 7.4.2), criteria for solid cantilevered shear
walls are expressed in the same context as for individual
piers between openings in a perforated shear wall.
C7.4.4.1 Stiffness
A. Linear Elastic Stiffness
Before initial cracking, behavior of reinforced wall or
pier components is essentially the same as for
unreinforced components, because the reinforcing steel
is strained at very low levels and the effective area of
masonry in tension is usually quite large relative to that
of the reinforcing bars. In this range, lateral stiffness of
wall or pier components may be determined assuming a
linear elastic analysis of components comprising
homogeneous materials. Equations C7-1 and C7-2 may
be used to determine lateral stiffness of walls and piers,
respectively, based on gross uncracked sections and
expected elastic moduli of masonry.
For a wall or pier component with sufficient shear
strength, flexural cracking will commence at lateral
force levels that are a fraction of the ultimate strength.
The fraction will depend on the relative amounts of
vertical reinforcement and masonry, the reinforcement
yield stress, the masonry compressive strength, the
length-to-height aspect ratio of the component, and the
amount of vertical compressive force. As a result of
flexural cracking, the lateral stiffness will reduce, since
the masonry is no longer effective in tension. This
reduction in stiffness will, however, result in an
essentially linear-elastic behavior, provided that the
masonry compressive stress remains at approximately
one half or less of the ultimate strength and the
reinforcement does not yield. Thus, lateral stiffness may
be represented with a reduced value representing the
effective cracked section.
FEMA 274 Seismic Rehabilitation Commentary
B. Nonlinear Behavior of Reinforced Masonry Walls
and Piers
Reinforced walls are known to soften when cracks
initiate. Vertical reinforcement becomes effective after
flexural cracks develop along mortar bed joints. With
further increase in lateral force, the vertical
reinforcement may yield, provided that adequate shear
strength is provided. The yielding steel will dissipate
substantial seismic energy. In such case, inelastic
deflection capacity will be limited by the ultimate
compressive strain in the masonry at the wall toe as the
steel strains reach well beyond their proportional limit.
Upon unloading, wall stresses will be relieved, but
deflections will not reduce substantially because cracks
will remain open. When force is reversed in direction,
the closing of previously opened cracks will be
restrained by the reinforcement acting in compression.
In this stage, the resistance of the section is primarily
from the reinforcement, and the stiffness will reduce
suddenly when the load is reversed. When cracks close
fully, the element stiffens, and resumes its character
from the loading portion of the previous half cycle. The
closing of cracks in the load reversal region causes a
“pinching” of the hysteretic loop, which reduces the
amount of energy dissipation, and increases the element
flexibility. After the first large-amplitude cycle,
conventional principles of mechanics used for elements
subjected to monotonically increasing loadings cannot
be used, because deformations in the masonry and the
steel, and at their interface, cannot be estimated reliably.
Approximate methods must be used to estimate
stiffness and deflection capacity.
Nonlinear behavior of RM wall components has been
studied, with large-scale experiments done on: (1)
single story walls (Shing et al., 1991), (2) two-story
walls (Merryman et al., 1990; Leiva and Klingner,
1991), and (3) a five-story building (Seible et al., 1994).
Dynamic testing of reduced-scale, reinforced concrete
masonry shear wall buildings by Paulson and Abrams
(1990) revealed substantial ductility and inelastic
energy dissipation.
C. Lateral Stiffness with Linear Procedures
The stiffness of RM wall and pier components that are
cracked can be an order of magnitude less than those
components that are uncracked. Because the length of
masonry walls in typical buildings can vary, some walls
are likely to crack while others remain uncracked.
Therefore, lateral stiffnesses should be based on the
consideration of whether individual components will
7-19
 Chapter 7: Masonry (Systematic Rehabilitation)

crack or not when subjected to expected amounts of where:

vertical and lateral force. This distinction is important
when: (1) distributing story shear force to individual fa = Expected amount of vertical compressive
walls, or shear force to adjacent piers in a perforated stress based on load combinations given in
shear wall, (2) estimating nonlinear force-deflection Equations 3-1 and 3-2
relations for wall or pier components with the Nonlinear fme = Expected masonry compressive strength as
Static or Dynamic Procedures, or (3) determining determined per Section 7.3.2.1
spectral accelerations based on periods of vibration for fye = Expected reinforcement yield stress as
the Linear Dynamic Procedure. determined per Section 7.3.2.6
The following criteria may be used to determine the heff = Height to resultant of lateral force
uncracked or cracked condition states as stated in L = Wall or pier length
Section 7.4.4.1.
Using Equation C7-8, the effective moment of inertia
if Q UF < M cr then I = I g (C7-5) can be determined without considering the amount of
lateral force or extent of cracking. This simplification
if Q UF ≥ M cr then I = I e (C7-6) avoids any iterations related to the interaction of
demand forces and stiffnesses—a cumbersome process,
particularly for deformation-controlled elements where
where: the elastic demand forces, QE, are fictitious, as
discussed in Section C7.4.2.1C. The derivation for
M cr = f te S g (C7-7) Equation C7-8 can be found in Priestley and Hart
(1989).
and:
C7.4.4.2 Strength Acceptance Criteria for
fte = Expected masonry tensile strength per Reinforced Masonry
Section 7.3.2.3
The requirements of Sections 7.4.4.2A, 7.4.4.2B, and
Ie = Effective moment of inertia based on 7.4.4.2C are based on the latest revisions to the NEHRP
cracking Recommended Seismic Provisions for New Buildings
Ig = Moment of inertia based on the uncracked (BSSC, 1995) for design of newly constructed
net mortared/grouted section reinforced masonry shear walls. The same assumptions,
QUF = Estimate of the maximum lateral force that procedures, and requirements are intended for existing
can be delivered to the component as defined wall or pier components.
with Equation 3-15
The lateral strength of RM wall or pier components is
Sg = Section modulus for the uncracked net governed by either flexural or shear action. The
mortared/grouted section
ultimate limit state for flexural action is masonry
compressive strain at the wall toe, or tensile fracture of
The stiffness of a cracked reinforced component can be vertical reinforcement. Shear strength is limited by
determined based on a moment-curvature analysis of a yielding of horizontal shear reinforcement, which
particular wall or pier cross section, recognizing the causes diagonal tension cracks to widen and, in so
amount and placement of vertical reinforcement, the doing, reduces aggregate interlock mechanisms. A
relative elastic moduli for the masonry and flexural mechanism should be considered as a
reinforcement, and the expected amounts of axial force deformation-controlled action because it involves
and bending moment. Alternatively, the secant stiffness yielding of reinforcement and some significant levels of
of a cracked reinforced component can be determined inelastic deformation capacity. Assumptions and
using Equation C7-8. procedures for determining expected lateral strength of
RM shear walls are given in Section 7.4.4.2A for
flexure.
I
---e- = 15, 000-
-----------------
fa
-------
1
-----------------------------------------
- (C7-8)
+
Ig f ye f me 2
1 + 0.75 ( L/h eff )

7-20 Seismic Rehabilitation Commentary FEMA 274

 Chapter 7: Masonry (Systematic Rehabilitation)
A shear mechanism should be considered as a force-
controlled action because it involves diagonal tension of
masonry. Assumptions and procedures for determining
the lower bound lateral strength of RM shear walls are
given in Section 7.4.4.2B.
The resistance of RM walls to vertical compressive
stress should be considered as a force-controlled action,
and should be characterized by the lower bound
strength given in Section 7.4.4.2D.
A. Expected Flexural Strength of Walls and Piers
Expected flexural strength of wall or pier components
shall be based on assumptions given in this section,
which are similar to those used for strength design of
reinforced concrete.
B. Lower Bound Shear Strength of Walls and Piers
Lower bound shear strength of RM wall or pier
components is limited to values given by Equations 7-9
and 7-10 for different moment-to-shear ratios. The
expected value of masonry compressive strength shall
be used to determine these limiting shear forces, which
are also considered to be expected values.
Shear resistance is assumed attributable to the strength
of both the masonry and reinforcement.
The previous criteria in the NEHRP Recommended
Provisions (BSSC, 1995) for shear in a plastic hinge
zone have been waived, since Equation 7-11 for
masonry shear strength is based on tests of shear walls
(Shing et al., 1991) where the shear was transferred
across a plastic hinge zone. Expected masonry
compressive strength, fme, and expected axial
compressive force, PCE, are to be used to determine the
expected masonry shear strength.
The lower bound shear strength attributable to the
horizontal reinforcement is given by Equation 7-12.
The previous form of this equation in the NEHRP
Provisions (BSSC, 1995) has been revised for clarity to
the more familiar format used for concrete members.
The limit that dv not exceed the wall height is intended
for squat walls (where dv is larger than h), so that the
assumed number of horizontal bars crossing a 45-
degree diagonal crack will not exceed the actual number
of bars. The 0.5 factor on reinforcement shear strength
is taken from research on reinforced masonry shear
walls (Shing et al., 1991) and accounts for nonuniform
FEMA 274 Seismic Rehabilitation Commentary
straining of horizontal reinforcement along the
component height.
C. Strength Considerations for Flanged Walls
Flanges on masonry shear walls will increase the lateral
strength and stiffness appreciably; however, they can
only be considered effective when the conditions of this
section are met.
The width of flange that may be considered effective in
compression or tension is based on research done on
reinforced masonry flanged walls (He and Priestley,
1992).
D. Lower Bound Vertical Compressive Strength of
Walls and Piers
Equation 7-13 for lower bound axial compressive
strength is similar to that for reinforced concrete
columns. Lower bound strengths of masonry and
reinforcement shall be used, rather than expected
strengths. The 0.8 factor represents a minimum
eccentricity of the vertical compressive load.
C7.4.4.3 Deformation Acceptance Criteria
A. Linear Procedures
For the Linear Static Procedure, m factors are given for
primary and secondary components for each
Performance Level in Table 7-4. Factors are given to
represent variable amounts of inelastic deformation
capacity for (1) various ratios of vertical compressive
stress to expected masonry compressive strength,
(2) wall or pier aspect ratios, and (3) index values
representing amounts of reinforcement, expected yield
stress of reinforcement, and expected masonry
compressive strength.
The m factors were determined from an analysis of
lateral deflections for reinforced wall or pier elements
based on the three parameters included in the table.
Curvature ductilities, µφ, were determined by dividing
the ultimate curvature, φu , by the curvature at first
yield, φy, per Equation C7-9.
φuMy
µ φ = ------------
- (C7-9)
φyMu
Displacement ductilities, µ∆, were then determined
from curvature ductilities, considering plastic rotations
7-21
 Chapter 7: Masonry (Systematic Rehabilitation)
at the base of component being limited to a plastic-
hinge zone length, lp, equal to:
l p = 0.2L + 0.04h eff
which then gave:
lp lp
µ ∆ = 1 + 3 ( µ φ – 1 )  ----  1 – 0.5 ----
L L
Analytical procedures were based on those presented in
Paulay and Priestley (1992).
For the Collapse Prevention Performance Level, m
factors were assigned equal to these displacement
ductilities.
Variable m factors are given for each Performance
Level, corresponding to approximate inelastic
deflections associated with specific damage states. For
Immediate Occupancy, some cracking can be tolerated
for typical occupancy conditions; m factors range from
1.0 to 4.0, depending on the amount of vertical
compressive stress, the aspect ratio, and the amount of
reinforcement. The Life Safety Performance Level is
related to lateral deflections associated with the
dislodgment of masonry units and/or severe cracking; m
factors are approximately twice those for Immediate
Occupancy. The Collapse Prevention Performance
Level is related to a loss of lateral strength for primary
components, and unstable gravity-load behavior of
secondary components; m factors are approximately
one-third larger than for Life Safety.
B. Nonlinear Procedures
Nonlinear deformation capacities for primary and
secondary components are represented in Figure 7-1
with dimensions d and e, respectively. These values are
consistent with the m values defined for each
Performance Level in Table 7-4.
Some cracking can be tolerated for Immediate
Occupancy. Because of the presence of reinforcement,
propagation of cracks will be limited, and thus
acceptable wall or pier drifts are larger than those for
URM walls.
7-22 Seismic Rehabilitation Commentary
The Life Safety Performance Level corresponds to
severe cracking of the masonry, or a potential for
masonry units to dislodge. If spacings of vertical and
horizontal reinforcement are equal to or less than
16 inches, these effects will be minimized, and the
(C7-10) acceptable drifts contained in Table 7-5 may be
increased by 25%.
Severe loss of lateral strength of a wall or pier element
can precipitate collapse of a lateral-load or gravity-load
structural system. In laboratory experiments, severe loss
(C7-11) of strength for in-plane reinforced masonry walls has
been observed to occur at lateral drifts exceeding 1.0%
for moderate amounts of reinforcement and vertical
compressive stress.
C7.4.5 RM Out-of-Plane Walls
Walls resisting lateral forces normal to their plane are
termed “out-of-plane walls.” The stiffness of walls
bending about their weak axis is three or more orders of
magnitude less than the stiffness of walls bending about
their strong axis. If a building system contains walls in
both directions, the stiffness of the transverse walls will
be insignificant. Analysis of out-of-plane walls with the
LSP is not warranted, because out-of-plane walls will
not attract appreciable lateral forces. Rather than design
on the basis of a pseudo lateral load applied to the
global structural system (as in Equation 3-6 with the
LSP), out-of-plane walls should resist inertial forces
that are prescribed in Section 2.11.7. For similar
reasons, the NSP is also not applicable for out-of-plane
walls. However, the Nonlinear Dynamic Procedure may
be useful for out-of-plane walls not complying with
strength criteria based on an equivalent static uniform
loading.
C7.4.5.1 Stiffness
The static behavior and dynamic response of RM walls
bending out-of-plane have revealed very large
flexibilities and inelastic deformation capacities.
Testing of wall panels is reported by Agbabian et al.
(1989), Hamid et al. (1989), and Blondet and Mayes
(1991). The effect of flexural cracking on stiffness is
quite significant, particularly for small percentages of
vertical reinforcement. The stiffness of a cracked
section can be as low as one-tenth that of the uncracked
section.
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
C7.4.5.2 Strength Acceptance Criteria
The strength of reinforced out-of-plane walls is nearly
always limited by flexural strength, because the span-
to-depth ratio is large.
Reinforced masonry walls usually have a single layer of
vertical reinforcement that is centered about a single
wythe for hollow-unit masonry, or between two wythes
of solid masonry. Nominal ultimate flexural capacity
can be calculated assuming a rectangular stress block
for the masonry in compression, which results in
Equation C7-12 for a section with a single layer of
tensile reinforcement.
 f ye 
Q CE = M CE = A s f ye d  1 – 0.59ρ ------- (C7-12)
 f me
Tests of RM walls have demonstrated the large inelastic
deformation capacity of wall panels subjected to out-of-
plane loadings. Deformation capacity is dependent on
the amount of vertical reinforcement, the level of
vertical compressive stress, and the height-to-thickness
aspect ratio.
C7.4.5.3 Deformation Acceptance Criteria
Out-of-plane RM walls can resist transverse inertial
loadings past the yield limit state with substantial
inelastic deformation capacity. If sufficient flexural
strength is available to resist the uniform face loading
prescribed in Section 2.11.7, and walls are tied to
diaphragms at their top and bottom, then they should
perform adequately for any level from Immediate
Occupancy to Collapse Prevention. Thus, no
performance limits are given on out-of-plane deflection
of wall panels since post-yield behavior will not need to
be relied on.
If the NDP is used, out-of-plane response of the
transverse walls may be determined for wall panels
performing in the nonlinear range of response. Whereas
the out-of-plane walls do not necessarily have to be
modeled as part of the global system if strength
requirements are met per Section C7.4.3.2, there is no
restriction excluding them from a model. On the
contrary, inclusion of the out-of-plane walls in a NDP
model may be necessary to demonstrate performance
for overly slender or weak walls. In such cases,
Performance Levels need to be defined in accordance
with the estimated out-of-plane deflection of the
transverse walls. Because out-of-plane masonry walls
FEMA 274 Seismic Rehabilitation Commentary
are local elements spanning across individual stories or
bays, the limit states in the following paragraphs are
expressed in terms of lateral deflection across their
story height or length between columns or pilasters.
Flexural cracking of an RM wall subjected to out-of-
plane bending should occur at the same drift level as for
an unreinforced wall. However, this will not, in general,
be associated with any Performance Level because
cracking of reinforced components is acceptable. As the
reinforcement yields at a story drift ratio of
approximately 2.0%, cracks will widen substantially
and may limit the immediate use of a building.
Life Safety is related to a wall panel reaching its peak
strength. This limit state has been estimated to occur at
a story drift ratio of 3%, based on experimental
research.
The loss of an entire out-of-plane wall may not
influence the integrity of the global structural system in
the direction under consideration. Therefore, the
Collapse Prevention Performance Level should not be
applicable for out-of-plane walls. However, the loss of
an out-of-plane wall will affect performance of the
system in the orthogonal direction when it acts as an in-
plane wall. Furthermore, loss of a wall panel can
seriously diminish the integrity of the gravity load
system if the wall is a bearing wall. Reinforced masonry
walls bending out-of-plane are very ductile. Collapse
should not occur unless lateral story drift ratios are very
large at 5% of the span or larger.
C7.5 Engineering Properties of
Masonry Infills
Masonry infill panels are found in most existing steel or
concrete frame building systems. Although they are a
result of architectural function, infill panels do resist
lateral forces with substantial structural action, and
should, therefore, be assumed to be part of the primary
lateral-force-resisting system.
Since infill panels are usually placed after floors are
constructed, they do not resist gravity dead loads at the
time of construction. However, if an infill is in tight
contact with the beam above, the panel may help
support live loads as well as dead loads from upper
stories if they are placed after installation of lower-level
infills. In addition, if the masonry infill materials tend to
expand with time (as is the case with some clay-unit
7-23
 Chapter 7: Masonry (Systematic Rehabilitation)
masonry), and/or the frame columns tend to shrink or
creep (the case with concrete columns), an infill panel
can attract vertical compressive stress as a portion of the
gravity loads are redistributed to it from the frame.
In Section 7.5, infill panels are not considered as
secondary elements even if they may support gravity
loads, because loss of an infill panel should not
jeopardize the vertical-load-carrying system. Typically,
frames are designed to resist 100% of gravity forces,
and should not suffer a loss in structural integrity if the
infill panels are eliminated.
If an infill panel is destroyed during seismic shaking,
and falls out from the surrounding frame, collapse of
the structural system can still be prevented, assuming
that the frame resists the full lateral load. If a lateral-
force analysis of the bare frame system demonstrates
prevention of collapse, then the infill panels should not
be subject to limits set forth by the Collapse Prevention
Performance Level.
C7.5.1 Types of Masonry Infills
The engineering properties given in Section 7.5 are
applicable to building systems with existing, enhanced,
or new masonry infills that combine to rehabilitate a
building system. In addition, the Guidelines provide
specific recommendations on minimum requirements
for enhancement of existing infill panels, in order that
their structural properties may be considered the same
as new or existing elements.
Stiffness assumptions, strength criteria, and acceptable
deflections for various limit states as described in
Sections 7.5.2 through 7.5.3 are common for existing or
enhanced masonry infills, or new masonry infills added
to an existing building system. Principles of mechanics
are the same regardless of the age of a masonry
element. Physically, there should be no difference in
stiffness assumptions, strength criteria, or inelastic
behavior for existing, enhanced, or newly constructed
infills. Thus, guidelines on determining engineering
properties for each of the three fundamental infill types
are expressed in common in these sections.
In Sections 7.5.2 through 7.5.3, infill panels subjected
to in-plane lateral forces are separated from walls
subjected to out-of-plane forces, because their
stiffnesses, strengths, and acceptable deformations are
quite different. Unreinforced masonry infills are
considered since they are the most common. However,
RM infills can be considered with the same criteria,
7-24 Seismic Rehabilitation Commentary
since the in-plane and out-of-plane mechanisms are not
influenced negatively by reinforcement.
C7.5.1.1 Existing Masonry Infills
Existing masonry infills will have a significant
influence on the lateral strength and drift of a building
system. Certain masonry infills may have a brittle
character; their removal may improve the overall
energy dissipation capabilities of a system, and thus be
an acceptable rehabilitation option. When considering a
particular rehabilitation scheme, existing masonry
infills, or their extraction, should be included in the
structural analysis along with any new masonry infill
panels that may be added.
A thorough condition assessment should be made of
existing masonry infills to increase the level of
confidence in characterizing structural properties.
Infilled frame buildings are mostly mid- to high-rise
buildings with steel or concrete gravity-load-resisting
systems and masonry infill perimeter walls. Steel frame
elements are often encased in concrete, brick, or tile for
fire protection purposes. For fire protection, masonry
infills may also be found within the interior of
buildings. Interior infills may extend up to the bottom
of beams or slabs, or they may stop at the ceiling level.
Floor framing systems in infilled buildings may consist
of almost any material. Because infilled frames tend to
be significantly stiffer than noninfilled frames, they are
likely to be the main lateral-force-resisting elements of
the building.
Typical masonry units used for infill panels are clay
bricks, concrete blocks, or hollow clay tile. For
buildings constructed earlier in this century, masonry
units were typically red clay bricks laid in lime mortar.
In more recent times, other types of units may have
been used, and mortars may have included portland or
masonry cement.
Clay-unit infills are common in two or three wythes,
and are bonded with headers every five to seven
courses. In many cases, the exterior wythe consists of a
facing of bricks, decorative terra cotta units, or cast
stone (or some combination of these) placed outside the
plane of the frame for architectural and weathering
purposes. In these cases, the brick wythe is attached to
the infill backing with intermittent header bricks or
corrugated metal ties placed in the mortar joints. Terra
cotta and stone veneers are typically anchored to the
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
infill backing with round metal tie rods bent in the form
of staples or hooks.
Location of the infill varies relative to the frame and the
connections between infills and frames. Commonly, the
interior wythes are supported on top of the beams and
the veneer masonry wythe is supported on a steel ledger
plate or angle cantilevering out from the beams. In other
cases, the outer wythe is supported by the keying action
of header bricks interlocked with the interior wythes.
The masonry units may be tightly fitted with the
surrounding frame units, or gaps may exist between the
frame and the infill.
Masonry infills may entirely fill one or more bays and
stories in a frame, although this condition is likely only
in walls away from the street. More commonly,
masonry infills are partial-height infills, or full-height
infills with window openings.
Infilled reinforced concrete or steel frames were
typically designed to carry all gravity loads and the
infills were not intended to be load bearing. Frames
were usually not designed for any significant lateral
loads. In reinforced concrete frames, beam
reinforcement is likely to not be continuous through the
joints, and the column bar splices may not be adequate
for tension forces. Frame elements may have some
widely spaced ties that are not likely to provide
adequate shear capacity or ductility.
Steel frames are commonly constructed with rolled
shapes for the lighter framing and riveted built-up
sections for the heavier framing. Beam connections are
usually semi-rigid, with beam seats and clip angles
connecting the beam flanges to the column. In some
cases, connections with gusset plates may have been
used in exterior frames to resist wind loads.
Infilled frames combine nonductile frame systems with
brittle masonry materials; hence they conceptually form
a poor lateral-load-resisting system. However,
observations from the 1906 San Francisco earthquake
and other subsequent earthquakes indicate a
surprisingly good performance for steel infilled frame
buildings. This good performance is attributed to (1) the
interaction of the infill with the steel frame, in which
the infill provides a significant bracing mechanism for
the frame, and (2) the fact that the steel frame members
possess adequate ductility to accommodate the demands
imposed on them by the infill. In addition, cracking of
the infill and the friction between the infill and the
FEMA 274 Seismic Rehabilitation Commentary
frame provides a significant energy dissipation
mechanism.
Reinforced concrete infilled frames have not fared as
well as steel infilled frames in severe earthquakes,
primarily due to the inability of nonductile concrete
members to accommodate the demands imposed on
them through the interaction with the infill.
Structural frame and masonry infill respond to lateral
shaking as a system, both frame and infill participating
in the response through a complex interaction. The
overall system response and the interaction between
frame and infill are influenced by the material and
geometric characteristics of each of these elements and
the variation of the element characteristics during
earthquake response.
The arrangement of infill panels along the height of the
building and in plan may have significant influence on
the overall earthquake response of the building. This
occurs, for example, when framing is kept open at the
street side of a building but is infilled along other
exterior frames. In this situation, there is the possibility
that the resulting asymmetry will produce increased
damage due to torsional response of the building.
Another case is the lack of infills at a lower story level,
which can result in an undesirable soft-story
configuration. Similar eccentric or soft-story conditions
may be created during the earthquake if infills in a
lower story and/or along a side of the building fail,
while infill panels in other locations remain relatively
undamaged. These overall system concerns can be
identified and considered in design if the response
behavior of the frame-infill system can be understood
and analyzed at the local, single infill panel level.
The failure modes of interest for earthquake
performance are as follows.
A. Dislodgment of Masonry Units During an
Earthquake
This may result from excessive deformations of the
infills due to in-plane or out-of-plane forces, or from
inadequate anchorage of veneer courses to the backing
courses. Where an exterior wythe of masonry extends
beyond the structural frame, delamination or splitting at
the collar joint may occur under the action of in-plane
loads. Because partial infills and infills with openings
are more flexible than solid infills, they may be more
prone to this type of damage.
7-25
 Chapter 7: Masonry (Systematic Rehabilitation)
B. Falling of Infill Panels
Infill panels (or large portions of wall) may fall out of
the surrounding frame due to inadequate out-of-plane
restraint at the frame-infill interface, or due to out-of-
plane flexural or shear failure of the infill panel. In
undamaged infills, these failures may result from out-
of-plane inertial forces, especially for infills at higher
story levels and with a large h/t ratio. However, it is
more likely for out-of-plane failure to occur after the
masonry units become dislodged due to damage from
in-plane loading.
C. In-Plane Failure of Infill Panels
Infill panels may lose their strength and stiffness due to
in-plane forces imparted to them during earthquake
response. This failure mode does not necessarily lead to
failure of the overall structural system, although the
changes in the strength and stiffnesses of the infill
panels are likely to have significant impact on the
overall structural response. Also, dislodgment of
masonry units or falling of infill panels are likely to
follow the failure of the infills due to in-plane
deformations. Shear strength of the infilled frame under
these circumstances would be expected to be controlled
by the shear capacity of the infill. Either of two modes
of failure may occur: sliding shear failure along a bed-
joint line (commonly about mid-height), or failure in
compression of the diagonal strut that forms within the
panel.
D. Premature Failure of Frame Elements or
Connections
The interaction of the frame with the infill during
earthquake shaking results in transfer of interactive
forces between frame members and the infill at contact
areas. These contact forces may generate internal forces
in frame members that are significantly different than
those determined by considering lateral response of the
frame alone (which has been the usual design
assumption in the past). Hence, premature failures may
occur in the beams, columns, or connections of the
frame. Examples of this behavior are the shear failures
induced in columns due to reduced effective flexural
length—which may occur when masonry infills form
only the spandrels above and below continuous window
openings (“captive columns”)—and failures of
columns, beams, and connections due to compressive
“strut” reactions imparted to them by the masonry infill.
Another mode of failure of frame elements is the failure
of the tension or compression chords of the infill frame
acting as a monolithic flexural element. This mode may
7-26 Seismic Rehabilitation Commentary
predominate in cases where the infill frame is relatively
slender and, in particular, where a single bay is infilled
in a multibay, multistory building. In this case, the infill
frame may act effectively as a flexurally controlled
shear wall, with the infill acting as the web and the
boundary columns acting as tension and compression
chords. Strength in this mode is calculated by
conventional flexural procedures, considering the
possibility of failure of either the tension chord or the
compression chord. Due consideration should be given
to tension chord splices, and to tension chord and
compression chord offset bars.
E. Failure of the Frame
Upon complete failure of the infill system—provided
that no premature failure of the frame elements has
occurred—the structural response and performance are
determined by the characteristics of the frame only
(except, perhaps, for the contribution of the damaged
infills to structural damping). As noted above, falling
infills present a hazard in themselves, and may also
produce a fundamental change in the response of the
infill structure. The response of the frame with the infill
missing should be assessed, keeping in mind the
likelihood that a soft story configuration or stiffness
eccentricity may have resulted.
C7.5.1.2 New Masonry Infills
Newly constructed masonry infill panels can be added
to an existing building system for the purpose of
strengthening, stiffening, or increasing inelastic
deformation and energy dissipation capacity.
Design of newly constructed masonry infill panels is
not addressed by any existing standards. Procedures for
estimating strength and stiffness for new infills shall be
in accordance with Sections 7.5.2 and 7.5.3.
C7.5.1.3 Enhanced Masonry Infills
Rehabilitation methods for masonry walls as described
in Section 7.4.1.3 are generally applicable as well for
masonry infills. In-plane strength and stiffness of a
perforated infill panel can be increased by infilling
openings with masonry, by applying shotcrete or
surface coatings to the face of an infill panel, by
injecting grout into the joints, or by repointing mortar
joints. Out-of-plane strength can be enhanced with
these methods in addition to providing stiffening
elements. Enlarging openings is not feasible for an infill
panel because panels elements are not susceptible to
rocking motions as are masonry piers or walls.
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
Reinforced or prestressed cores are not practical
because vertical coring of an infill panel is difficult.
In addition, the following two enhancement methods
are unique to infill rehabilitation.
A. Boundary Restraints for Infill Panels
The stability of isolated infill panels with gaps between
them and the surrounding frame may be improved by
restraining out-of-plane movements with steel fixtures
that are anchored to the adjacent frame members. This
method does not fill in the gaps, and therefore does not
improve in-plane action.
B. Joints Around Infill Panels
Infill panels with gaps around their perimeter do not
fully participate in resisting lateral forces. Furthermore,
such walls require perimeter restraints for out-of-plane
forces. By filling gaps around an infill panel, multiple
benefits can be gained, including increased in-plane
strength and stiffness, increased out-of-plane strength
(through arching action), and elimination of the need
for out-of-plane perimeter restraints.
C7.5.2 In-Plane Masonry Infills
Infill panels resisting lateral forces parallel to their
plane are termed “in-plane infills.”
Behavior of infilled frame systems subjected to in-plane
lateral forces is influenced by mechanical properties of
both the frame and infill materials, stress or lateral
deformation levels, existence of openings in the infill,
and the geometrical proportions of the system.
Existence of an initial gap between the frame members
and the infill also influences the behavior of the system.
C7.5.2.1 Stiffness
In-plane lateral stiffness of an infilled frame system is
not the same as the sum of the frame and infill
stiffnesses, because of the interaction of the infill with
the surrounding frame. Experiments have shown that
under lateral forces, the frame tends to separate from
the infill near windward lower and leeward upper
corners of the infill panels, causing compressive contact
stresses to develop between the frame and the infill at
the other diagonally opposite corners. Recognizing this
behavior, the stiffness contribution of the infill is
represented with an equivalent compression strut
connecting windward upper and leeward lower corners
of the infilled frame. In such an analytical model, if the
FEMA 274 Seismic Rehabilitation Commentary
thickness and modulus of elasticity of the strut are
assumed to be the same as those of the infill, the
problem is reduced to determining the effective width
of the compression strut. Solidly infilled frames may be
modeled with a single compression strut in this fashion.
For global building analysis purposes, the compression
struts representing infill stiffness of solid infill panels
may be placed concentrically across the diagonals of the
frame, effectively forming a concentrically braced
frame system (Figure C7-1). In this configuration,
however, the forces imposed on columns (and beams)
of the frame by the infill are not represented. To account
for these effects, compression struts may be placed
eccentrically within the frames as shown in
Figure C7-2. If the analytical models incorporate
eccentrically located compression struts, the results
should yield infill effects on columns directly.
Alternatively, global analyses may be performed using
concentric braced frame models, and the infill effects on
columns (or beams) may be evaluated at a local level by
applying the strut loads onto the columns (or beams).
Diagonally concentric equivalent struts may also be
used to incorporate infill panel stiffnesses into
analytical models for perforated infill panels (e.g.,
infills with window openings), provided that the
equivalent stiffness of the infill is determined using
appropriate analysis methods (e.g., finite element
analysis) in a consistent fashion with the global
analytical model. Analysis of local effects, however,
must consider various possible stress fields that can
potentially develop within the infill. A possible
representation of these stress fields with multiple
compression struts, as shown in Figure C7-3, have been
proposed by Hamburger (1993). Theoretical work and
experimental data for determining multiple strut
placement and strut properties, however, are not
sufficient to establish reliable guidelines; the use of this
approach requires exercise of judgment on a case-by-
case basis.
The equivalent strut concept was first proposed by
Polyakov (1960). Since then, Holmes (1961, 1963),
Stafford Smith (1962, 1966, 1968) Stafford Smith and
Carter (1969), Mainstone (1971 and 1974), Mainstone
and Weeks (1971), and others have proposed methods
and relationships to determine equivalent strut
properties. Klingner & Bertero (1976) have found the
method developed by Mainstone to provide reasonable
approximation to observed behavior of infill panels.
7-27
 ,,
Chapter 7: Masonry (Systematic Rehabilitation)
,,
,,
,,
rin
,,
Fxi
a hinf
Figure C7-1
,,
Compression Strut Analogy–Concentric Struts
Angel et al. (1994) have found a strut width equal to
one-eighth of the diagonal dimension of the infill panel
to provide good correlation with experimental results;
they also proposed modifications to the frame-infill
system stiffness expression developed by Holmes to
account for the effects of cyclic loading.
In addition to these empirical studies, frame infill
systems have been studied using detailed finite element
models (Lotfi and Shing, 1994; Durrani and Luo, 1994;
Mehrabi and Shing, 1994; Gergely et al., 1994; Kariotis
et al., 1994). Although it is not presently practical to use
general-purpose finite element software to perform
detailed nonlinear finite element analyses of infill
frames, recently developed special-purpose computer
software, such as FEM/I (Ewing et al., 1987) may be
used to determine equivalent strut properties from
nonlinear finite element analyses of typical frame-infill
configurations. With such special purpose software, the
7-28 Seismic Rehabilitation Commentary
Linf
f
hcol
force-deformation behavior of the frame-infill system is
determined through nonlinear finite element analysis,
and the equivalent strut properties for use in elastic
models are derived from the force-deformation
relationship for a target displacement.
Experimental studies done at the Y-12 Plant of the Oak
Ridge National Laboratory (Flanagan et al., 1994)
showed that the same equivalent strut modeling
procedures could be used for infill panels constructed
with hollow-clay tile.
In the Guidelines, the equivalent compression strut
model is adopted to represent the in-plane stiffness of
solid masonry infill panels. The relationship used to
determine the strut width, Equation 7-14, has been
proposed by Mainstone (1971). There are not sufficient
data to provide modeling guidelines for representing
stiffness of perforated infill wall panels with multiple
FEMA 274
 ,,, ,,,
Chapter 7: Masonry (Systematic Rehabilitation)

,,,
,,, ,,,
,,,
,,,
,, ,,,
Fxi Fxi

Figure C7-2
,, ,,,
,, ,,,
Compression Strut Analogy–Eccentric
Struts
Figure C7-3 Compression Strut Analogy–Perforated
Infills

equivalent struts. However, as discussed above,
equivalent struts may still be used in analyses of infilled
frames with perforated infills, provided that the
equivalent strut properties are derived from detailed
finite element analyses of representative frame-infill
systems.
C7.5.2.2 Strength Acceptance Criteria
A. Infill Shear Strength
The horizontal component of the force resisted by the
equivalent strut should be compared with the expected
shear strength of an infill panel times the appropriate m
and κ factors per the load combination given in
Equation 3-18.
The expected infill strength as given with
Equation 7-15 is based on an average shear stress across
the net mortared/grouted area of a horizontal section cut
across the panel. The expected shear strength across this
area, fvie, is taken as the expected bed-joint shear
strength, vme, for existing construction, or values based
on the 1994 NEHRP Recommended Provisions (BSSC,
1995) for new construction. No allowance is made for
shear strength enhancements due to vertical
compressive stress, because gravity forces are assumed
to be resisted by the frame.
The expected infill shear strength is based on bed-joint
sliding with no confinement from the surrounding
frame, and may thus be less than the actual shear
strength. A study done by Angel et. al. (1994) found
that results from in-place shear tests provide a
conservative estimate of infill shear strength. A
resolution based on discussions at an NCEER
Workshop on Masonry Infills (Abrams, ed., 1994) was
that average infill shear stress provided a good index of
lateral infill shear strength.

FEMA 274 Seismic Rehabilitation Commentary 7-29

 Chapter 7: Masonry (Systematic Rehabilitation)
B. Required Strength of Column Members Adjacent
to Infill Panels
Infill panels can attract substantial forces to adjacent
frame members. These forces can be more demanding
of the strength and inelastic deformation capacity of
beam and column members than those resulting from
lateral design forces applied to a bare frame. Because a
stiff masonry infill panel can attract more lateral force
than a frame can resist, frames must be checked to see if
they are capable of resisting infill forces in the ductile
manner that is assumed for their design or evaluation.
Shear strength of the column members should be
checked to resist either the horizontal component of the
axial force in equivalent struts, or the shear forces
resulting from development of plastic hinges at the top
and bottom of a column of reduced height. Although
neither of these two conditions is exactly representative
of what may occur—because of the complex
interactions between a frame and an infill panel—these
criteria should result in an adequate check to insure
ductility of the frame.
The first condition is depicted in Figure C7-4, where the
equivalent strut is assumed to be acting eccentrically
about the beam-to-column joint with the action
illustrated in Figure C7-2. For simplicity, the strut force
is assumed to be applied to the column member at the
edge of its equivalent width, a. This assumption results
in a short shear span of the column equal to lceff, for
which the horizontal strut component must be resisted
over. The infill force applied to the frame should be an
expected value and not an unreduced elastic demand
force as determined with the LSP. The strength of the
column member is also an expected strength. Thus, the
relative m factors for both the column and the infill
panel should be considered when checking the column
strength for this action.
Because the first condition can result in excessively
high column shear forces, a second option is based on
achieving ductile performance of the column when
partially braced by the infill panel. This second option
consists of checking column shear strength for resisting
expected flexural strengths applied at the top and
bottom of a short column portion of height lceff. This
requirement may lead to smaller shear forces for
relatively light column flexural strengths and will insure
that hinging of the column members will occur. The
same condition shall be applied to captive columns
braced with partial height infills.
7-30 Seismic Rehabilitation Commentary
Effects of infill panels on frames may be neglected if
the bed-joint shear strength of masonry is known to be
sufficiently low. In this case, the infill panel will
conform to the deflected shape of the frame as courses
of masonry slide relative to one another across bed
joints. The limit of 50 psi for expected masonry strength
defines this sufficiently weak condition, which must be
determined from in-place shear tests.
C. Required Strength of Beam Members Adjacent to
Infill Panels
For the same reasons as discussed for column members
adjacent to infill panels in the preceding section,
flexural and shear strengths of beam members must be
checked to ensure the transfer of eccentric infill vertical
force components. Again, two options are given to
check either strength or deformation capacity of the
beam. Equations 7-18 and 7-19 are based on the
geometry of forces as shown in Figure C7-5.
C7.5.2.3 Deformation Acceptance Criteria
A. Linear Procedures
In Table 7-6, m factors are given only for infill panels
acting as primary elements. Because the surrounding
frame is assumed to resist gravity forces, the only
structural role of the infill is to resist lateral forces,
which is a primary action. Thus, infill panels are not
considered to act as secondary members and do not
need to be checked for their ability to support gravity
loads while deflecting laterally.
No m factors are given in Table 7-6 for the Collapse
Prevention Performance Level because loss of an entire
infill panel should not result in collapse of the frame
system. In this case, component behavior is not related
to performance of the system. However, the ability of
the bare frame to resist gravity and lateral forces must
be checked to see if collapse will be prevented.
Amounts of inelastic deformation for an infill panel are
expressed in terms of a β factor that expresses the
relative frame to infill strength. When the expected
lateral frame strength exceeds approximately 1.3 times
the expected shear strength of an infill, any sudden loss
of infill strength is not likely to result in a substantial
decrease in lateral strength of the frame-infill system.
Furthermore, when the frame is strong relative to the
infill, it will offer more confinement to the infill
because inelastic deformations of frame members will
be minimized. When the expected strength of the frame
is approximately less than 0.7 times the infill expected
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
,,,,
,,,,
lceff = a
cos θc
Figure C7-4
hinf
,,,,
Estimating Forces Applied to Columns
strength, a sudden loss of infill strength may result in a
sudden and substantial decrease in strength of the
frame-infill system. Also, when the frame is weak
relative to the infill, confinement effects will be reduced
as inelastic deformations of frame members occur.
Inelastic deformation capacity of infills is also
expressed in terms of the length-to-height aspect ratio
of an infill panel. Larger m factors are given for more
slender panels than stocky panels because they will be
more flexible and thus more adaptable to frame
distortions. For taller panels, the angle of the equivalent
strut relative to the horizontal will be larger than for
stocky panels, and thus offer less resistance to lateral
forces.
For the Immediate Occupancy Performance Level,
some minor cracking of an infill panel is permissible,
and thus m factors in Table 7-6 are larger than one
inferring that some inelastic deformations can occur.
However, when the frame strength is low relative to that
of the infill, cracking of the infill can result in damage
to the adjacent frame, which could alter the
performance of the frame-infill system. Thus, for low β
values, the m values should be limited to 1.0. For
systems with moderate or large β values, no distinction
is made in m values for the relative frame-to-infill
FEMA 274 Seismic Rehabilitation Commentary
Linf
a
θc
strength because this level of infill should not result in
damage to frame members.
B. Nonlinear Procedures
In Table 7-7, inelastic deformation capacities of
masonry infill panels are expressed with the d
dimension, which is given in terms of the generalized
force-deflection relations as depicted in Figure 7-1. No
values for terms c or e are given in the table because
they apply only to secondary elements. For the reasons
discussed in the previous section, infill panels are
considered only as primary elements.
Deformation capacity and acceptable deformations are
expressed in terms of the relative frame-to-infill
strength and the panel aspect ratio, as is done with the m
factors in Table 7-6.
At a very low level of story drift ratio (on the order of
0.01%), the leeward column of an infilled frame will
separate from the infill, resulting in a sudden loss of
stiffness. This limit state is of little concern, since the
gap will not be visible following the earthquake, and the
analysis should have neglected any tension across the
gap by using a compression strut. For such a case, the
initial stiffness should be based on the axial stiffness of
the equivalent strut with properties as defined with
7-31
 Chapter 7: Masonry (Systematic Rehabilitation)
,,,,
lbeff = a Linf
sin θb
,,,,
,,,,
,,,,
hinf
Figure C7-5 Estimating Forces Applied to Beams
Equation 7-14, rather than on a fully uncracked solid
panel with full contact with the frame on all edges.
As the infill shear stress is increased, minor cracking
along bed joints will develop for weaker mortars, or
diagonal cracks will form across a panel for stronger
mortars. This will occur at a story drift ratio of
nominally 0.1% for square panels. Initial cracking of an
infill panel will result in a decreased stiffness, but the
panel will still continue to resist increased shear forces
if confined by the surrounding frame. Following an
earthquake, these minor cracks may be noticeable, but
no structural repair would be necessary.
Further loading will result in a wider dispersion of bed
joint cracks, or an elongation of diagonal cracks.
Moderate or severe cracking of a square masonry infill
panel can be expected at story drift ratio levels of
approximately 0.3% or more. Even in this condition, an
infill panel may continue to provide resistance if the
surrounding frame is in tight contact and can provide
confinement to the masonry assemblage.
Life Safety corresponds to reaching the peak infill
strength. In some cases, Life Safety may also be related
to dislodgment and falling of masonry units because of
the hazard to life or the blocking of egress. For this
7-32 Seismic Rehabilitation Commentary
a
θb
case, the relative frame-to-infill strength becomes a
significant parameter because post-cracked behavior of
a masonry infill panel is very much dependent on the
confinement offered by the frame.
Experimental studies done at the Y-12 Plant of the Oak
Ridge National Laboratory (Flanagan et al., 1993)
showed that the same force-deflection properties could
be used for infill panels constructed with hollow-clay
tile.
C7.5.3 Out-of-Plane Masonry Infills
Infill panels resisting lateral forces normal to their plane
are termed “out-of-plane infills.” The minimum height-
to-thickness ratios given in Table 7-8 are based on
achieving a transverse infill strength based on an
arching action model that will exceed any plausible
acceleration level for each of the various seismic zones.
C7.5.3.1 Stiffness
The stiffness of infill panels bending about their weak
axes is three or more orders of magnitude less than the
stiffness of panels bending about their strong axes.
Thus, in an analysis of a building system with infills or
walls in each direction, the stiffness of the transverse
infills can be neglected.
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
The out-of-plane deflection of an infill panel can be
approximated by considering strips of unit width
spanning either vertically between floors or horizontally
between columns. The uncracked stiffness of the strip
can be considered if the maximum bending moment is
less than the cracking moment. Post-cracked behavior
can be tolerated, provided that conditions exist for
arching action to take place.
The restrictions on when arching action can be
considered are based on the ability of the panel to
develop internal thrusts when being loaded transversely.
The panel must be in tight contact with the surrounding
beam and column members. These members must have
a flexural stiffness sufficiently high so that they will not
flex when subjected to the infill thrust forces, as well as
a flexural strength large enough to resist the thrusts.
Slender panels may snap through the frame, particularly
if ultimate masonry compressive strains are large at
their boundaries. Studies done by Angel et al. (1994)
have shown that this may occur for panels with a
hinf /tinf ratio exceeding 20 if the ultimate strain is
0.005. This slenderness has been set as a limit on when
arching action may be considered.
Transverse deflections at mid-length of a one-way strip
for panels that will not snap through the frame can be
determined with Equation 7-20, which is a simplified
version of an equation given by Abrams et al. (1993)
assuming arching action and an ultimate masonry
compressive strain equal to 0.004.
C7.5.3.2 Strength Acceptability Criteria
Out-of-plane infills should not be evaluated using the
Linear or Nonlinear Static Procedures of Chapter 3
because these infills act as isolated elements spanning
across individual stories. The transverse strength of
infill panels should exceed the maximum plausible
lateral inertial forces that result from the mass of the
panel accelerating. Because the evaluation of out-of-
plane infill panels does not depend on an unreduced
value of base shear—as is done for in-plane
components per the LSP—there is no need to use
expected values of strength. Thus, strength criteria
given in this section are based on lower bound estimates
of strength. Actual transverse strengths can be higher.
Masonry infill panels must be restrained perpendicular
to the wall surface on all four sides in order to prevent
the whole infill panel, or large portions of it, from
FEMA 274 Seismic Rehabilitation Commentary
sliding and falling outward. Exterior wythes of
multiwythe infills should be restrained from separating
or peeling from the interior wythe (see
Section C11.9.1.2A). Field and test observations
indicate that infills constructed in tight contact with the
surrounding frame can be considered to have adequate
out-of-plane restraint. If a gap exists between the frame
and the infill on any side, the gap must be filled with
grout to provide tight contact, or out-of-plane restraint
must be provided with other mechanical means.
Infills that are in tight contact with perimeter frame
members develop arching mechanisms when subjected
to out-of-plane loads. The out-of-plane capacity of an
infill panel can be increased substantially through such
an arching mechanism. However, formation of arching
mechanisms requires the frame members to have
substantial stiffness and strength to resist the thrust
forces imparted on them by the arching infill. In
general, if the infills are continuous—that is, adjacent
bays and story levels are also infilled—the boundary
conditions required for arch-mechanism formation may
be assumed to be satisfied. For infills with open
adjacent bays or story levels, the strength and stiffness
of the frame members must be checked to confirm their
adequacy.
A lower bound estimate of the transverse infill strength
is given by Equation 7-21. The equation is a simplified
version of one derived by Angel et al. (1994).
Flexibility of beam or column members is included in
the expression if their EfeIf values exceed the minimum
of 3.6 x 106 lb-in.2 as specified in the previous section.
According to the theory, frame members with
stiffnesses as low as this value should lower transverse
strength by as much as 0.6. The lower bound strength
equation also includes a reduction of 76% for an
estimated amount of in-plane cracking for the most
slender panel permitted. In this case, in-plane
deflections equal to 50% more than those at initial
cracking have been assumed.
C7.5.3.3 Deformation Acceptability Criteria
Because out-of-plane infills are local elements spanning
across individual stories and bays, limit states are
expressed in terms of lateral deflection across their
story height or length between columns.
The Immediate Occupancy Performance Level is not
necessarily related to initial cracking of a wall. Some
7-33
 Chapter 7: Masonry (Systematic Rehabilitation)

cracking can be tolerated for typical occupancy C7.7 Masonry Foundation Elements
conditions.
No commentary is provided for this section.
Life Safety is related to extensive cracking of the infill
panel. If arching action can be developed, the lateral
story drift ratio of the most slender panel permitted
C7.8 Definitions
(hinf /tinf = 20) according to Equation 7-20 will be All definitions for Chapter 7 are given in the
2.8%, which is just less than the limit of 3.0% given for Guidelines.
Life Safety. Thus, all infills that can develop arching
mechanisms can meet this required Performance Level,
provided that their strength will be sufficient to resist C7.9 Symbols
inertial forces.
Av Shear area of wall or pier, in.2
Eme Expected elastic modulus of masonry in
C7.6 Anchorage to Masonry Walls compression as determined in Section 7.3.2.2,
According to Section 8.3.12 of BSSC (1995), the psi
pullout strength of anchors is governed by the strength Gme Shear modulus of masonry as determined in
of the steel or the anchorage strength of the masonry. Section 7.3.2.5, psi
When practical, sufficient anchorage should be Ie Effective moment of inertia of reinforced wall or
provided so that the anchor steel will yield, and a brittle
pier per Equation C7-8, in.4
pullout failure will be avoided. A ductile anchor will
help insure a uniform distribution of force to individual Ig Moment of inertia for uncracked, gross section,
anchors in the case that one or a few anchors are in.4
overloaded. If Moment of inertia of beam or column member,
in.4
Ductility of an anchor will not significantly influence
global ductility of a structural system, because plastic L Length of wall or pier, in.
anchor extensions will be quite short relative to inelastic Linf Length of infill panel, in.
deformations of structural members. Anchors should be Mu Moment at crushing of masonry, lb-in.
considered as force-controlled components, to ensure
that the forces delivered to them by adjacent members My Moment at yield of reinforcement, lb-in.
will be resisted without inelastic straining or pullout of Q CE Lower-bound estimate of the strength of a
the anchor. component or element at the deformation level
under consideration
The effective embedment length is the length used to
Q Deformation-controlled design action
estimate the projected area of a pullout cone of UD
masonry. Per Section 8.3.12 of BSSC (1995), this R1 Out-of-plane infill strength reduction factor to
length is the length of embedment normal to the wall account for in-plane damage
surface to the bearing surface of an anchor plate or head
of an anchor bolt, or within one bar diameter from a a Width of equivalent strut representing in-plane
hooked end. infill panel, in.
When the embedment length is less than the minimum d Effective depth of reinforced section, in.
length prescribed by Section 8.3.12.1.4 of BSSC fa Expected amount of vertical compressive stress
(1995), the pullout strength cannot be estimated based on load combinations given in
reliably. Equations 3-1 and 3-2, psi
fme Expected compressive strength of masonry as
Shear strength of anchorages with edge distances less determined per Section 7.3.2.1, psi
than 12 bolt diameters can be reduced by linear fte Expected masonry tensile strength as
interpolation to zero at an embedment distance of one determined per Section 7.3.2.3, psi
inch (25.4 mm).

7-34 Seismic Rehabilitation Commentary FEMA 274

 Chapter 7: Masonry (Systematic Rehabilitation)
fye Expected yield strength of reinforcing steel as
determined per Section 7.3.2.6, psi
heff Height to resultant of lateral force for wall or
pier, in.
k Lateral stiffness of shear wall or pier, lb-in.
lbeff Assumed distance to infill strut reaction point
for beams as shown in Figure C7-5
lceff Assumed distance to infill strut reaction point
for columns as shown in Figure C7-4
lp Length of plastic hinge for reinforced masonry
wall or pier, in.
m Factor to account for inelastic deformation
capacity used in Equation 3-18
qcr Uniform transverse load when flexural cracking
commences
vt Wall shear strength, 50th percentile, psi
∆cr In-plane deflection of infill panel at first
cracking, in.
∆inf Out-of-plane deflection of infill panel at
midspan, in.
εmu Crushing strain of masonry
µ∆ Displacement ductility for reinforced wall or
pier section
µφ Curvature ductility for reinforced wall or pier
section
θb Angle between lower edge of compression strut
and beam as shown in Figure C7-5, radians
θc Angle between lower edge of compression strut
and beam as shown in Figure C7-4, radians
φy Curvature at initial yield of reinforcement, 1/in.
φu Curvature at crushing of masonry, 1/in.
C7.10 References
ABK, 1981, Methodology for Mitigation of Seismic
Hazards in Existing Unreinforced Masonry Buildings:
Wall Testing, Out-of-Plane, Topical Report 04.
Abrams, D. P., 1992, “Strength and Behavior of
Unreinforced Masonry Elements,” Proceedings of Tenth
World Conference on Earthquake Engineering,
Balkema Publishers, Rotterdam, The Netherlands,
pp. 3475–3480.
FEMA 274 Seismic Rehabilitation Commentary
Abrams, D. P., editor, 1994, Proceedings of the NCEER
Workshop on Seismic Response of Masonry Infills,
Report No. NCEER-94-0004, National Center for
Earthquake Engineering Research, State University of
New York at Buffalo, New York.
Abrams, D. P., and Matthys, J., 1991, “Present and
Future Techniques for Nondestructive Evaluation of
Masonry Structures,” Journal of The Masonry Society,
Boulder, Colorado, Vol. 10, No. 1, pp. 22–30.
Abrams, D. P., and Shah, N., 1992, Cyclic Load Testing
of Unreinforced Masonry Walls, Advanced
Construction Technology Center Report #92-26-10,
College of Engineering, University of Illinois at
Urbana, Illinois.
Abrams, D. P., Angel, R., and Uzarski, J., 1993,
“Transverse Strength of Damaged URM Infills,”
Journal of The Masonry Society, Boulder, Colorado,
Vol. 12, No. 1, pp. 45–52.
Agbabian, M. S., Adham, S., Masri, S., Avanessian, V.,
and Traina, I., 1989, Out-of-Plane Dynamic Testing of
Concrete Masonry Walls, TCCMAR Report No.
3.02(b1), University of Southern California, Los
Angeles, California.
Angel, R., Abrams, D. P., Shapiro, D., Uzarski, J., and
Webster, M., 1994, Behavior of Reinforced Concrete
Frames with Masonry Infills, Structural Research Series
No. 589, UILU-ENG-94-2005, University of Illinois at
Urbana, Illinois.
ASTM, latest edition, Standards with the following
numbers: C597, C1072, C1197, E518, E519, American
Society for Testing Materials, Philadelphia,
Pennsylvania.
Atkinson, R. H., Amadei, B. P., Saeb, S., and Sture, S.,
1989, “Response of Masonry Bed Joints in Direct
Shear,” Journal of the Structural Engineering Division,
American Society of Civil Engineers, New York, New
York, Vol. 115, No. 9, pp. 2276–2296.
Blondet, M., and Mayes, R. L., 1991, The Transverse
Response of Clay Masonry Walls Subjected to Strong
Motion Earthquakes, TCCMAR Report No. 3.2(b2),
Computech Engineering Services, Inc., Berkeley,
California.
7-35
 Chapter 7: Masonry (Systematic Rehabilitation)
BSSC, 1992, NEHRP Handbook for the Seismic
Evaluation of Existing Buildings, developed by the
Building Seismic Safety Council for the Federal
Emergency Management Agency (Report No.
FEMA 178), Washington, D.C.
BSSC, 1995, NEHRP Recommended Provisions for
Seismic Regulations for New Buildings, 1994 Edition,
Part 1: Provisions and Part 2: Commentary, prepared
by the Building Seismic Safety Council for the Federal
Emergency Management Agency (Report Nos. FEMA
222A and 223A), Washington, D.C.
Calvi, G. M., 1988, “Correlation between Ultrasonic
and Load Tests on Old Masonry Specimens,”
Proceedings of Eighth International Brick/Block
Masonry Conference, Elsevier Applied Science, Essex,
England, pp. 1665–1672.
Calvi, G. M., Kingsley, G. R., and Magenes, G., 1996,
“Testing of Masonry Structures for Seismic
Assessment,” Earthquake Spectra, Earthquake
Engineering Research Institute, Oakland, California,
Vol. 12, No. 1, pp. 145–162.
Costley, A. C., and Abrams, D. P., 1995, Dynamic
Response of URM Buildings with Flexible Diaphragms,
Structural Research Series No. 605, University of
Illinois at Urbana, Illinois.
Curtin, W. G., Shaw, G., and Beck, J. K., 1988, Design
of Reinforced and Prestressed Masonry, Thomas
Telford Publishers, London, United Kingdom.
Durrani, A. J., and Luo, Y. H., 1994, “Seismic Retrofit
of Flat-Slab Buildings with Masonry Infills,”
Proceedings of NCEER Workshop on Seismic Response
of Masonry Infills, Technical Report NCEER-94-0004,
San Francisco, California, National Center for
Earthquake Engineering Research, State University of
New York at Buffalo, New York.
Epperson, G. S., and Abrams, D. P., 1989,
Nondestructive Evaluation of Masonry Buildings,
Advanced Construction Technology Center Report No.
89-26-03, College of Engineering, University of Illinois
at Urbana, Illinois.
Epperson, G. S., and Abrams, D. P., 1992, “Evaluating
Lateral Strength of Existing Unreinforced Brick
Masonry Piers in the Laboratory,” Journal of The
7-36 Seismic Rehabilitation Commentary
Masonry Society, Boulder, Colorado, Vol. 10, No. 2,
pp. 86–93.
Ewing, R. D., El-Mustapha, A. M., and Kariotis, J. C.,
1987, FEM/I: A Finite Element Computer Program for
the Nonlinear Static Analysis of Reinforced Masonry
Building Components, TCCMAR Report No. 2.2-1,
R. D. Ewing and Associates, Palos Verdes, California.
Flanagan, R. D., Bennett, R. M., and Barclay, G. A.,
1993, “In-Plane Behavior and Strength of Structural
Clay Tile Infilled Frames,” Proceedings of the Sixth
North American Masonry Conference, Drexel
University, Philadelphia, Pennsylvania, pp. 371–382.
Flanagan, R. D., Tenbus, M. A., and Bennett, R. M.,
1994, “Numerical Modeling of Clay-Tile Infills,”
Proceedings of NCEER Workshop on Seismic Response
of Masonry Infills, Technical Report NCEER-94-0004,
San Francisco, California, National Center for
Earthquake Engineering Research, State University of
New York at Buffalo, New York.
Gergely, P., White, R. N., and Mosalam, K. M., 1994,
“Evaluation and Modeling of Infilled Frames,”
Proceedings of NCEER Workshop on Seismic Response
of Masonry Infills, Technical Report NCEER-94-0004,
San Francisco, California, National Center for
Earthquake Engineering Research, State University of
New York at Buffalo, New York.
Hamburger, R. O., 1993, “Methodology for Seismic
Capacity Evaluation of Steel-Frame Buildings with
Infill Unreinforced Masonry,” Proceedings of 1993
National Earthquake Conference, Central U.S.
Earthquake Consortium, Memphis, Tennessee, Vol. II,
pp. 173–191.
Hamid, A. A., Abboud, B. E., Farah, M. W.,
Hatem, M. K., and Harris, H. G., 1989, Response of
Reinforced Block Masonry Walls to Out-of-Plane Static
Loads, TCCMAR Report No. 3.2(a), Department of
Civil and Environmental Engineering, Drexel
University, Philadelphia, Pennsylvania.
He, L., and Priestley, M.J.N., 1992, Seismic Behavior of
Flanged Masonry Shear Walls, TCCMAR Report No.
4.1-2, Department of Applied Mechanics and
Engineering Sciences, University of California at San
Diego, California.
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)
Holmes, M., 1961, “Steel Frames with Brickwork and
Concrete Infilling,” Proceedings of Institution of Civil
Engineers, London, United Kingdom, Vol. 19,
pp. 473–478.
Holmes, M., 1963, “Combined Loading on Infilled
Frames,” Proceedings of Institution of Civil Engineers,
London, United Kingdom, Vol. 25, pp. 31–38.
ICBO, 1994, Uniform Code for Building Conservation,
International Conference of Building Officials,
Whittier, California.
Kariotis, J., Guh, T. J., Hill, J. A., and Youssef, N.F.G.,
1994, “Simulation of the Recorded Response of
Unreinforced (URM) Infill Buildings,” Proceedings of
NCEER Workshop on Seismic Response of Masonry
Infills, Technical Report NCEER-94-0004, San
Francisco, California, National Center for Earthquake
Engineering Research, State University of New York at
Buffalo, New York.
Kariotis, J., and Ngheim, D., 1995, “In-Situ
Determination of Compressive Stress-Strain
Relationship of Multi-Wythe Brick Masonry,”
Proceedings of ASCE Structures Congress, American
Society of Civil Engineers, New York, New York,
Vol. 2, pp. 1421-1426.
Kingsley, G. R., Noland, J. L., and Atkinson, R. H.,
1987, “Nondestructive Evaluation of Masonry
Structures Using Sonic and Ultrasonic Pulse Velocity
Techniques,” Proceedings of Fourth North American
Masonry Conference, The Masonry Society, Boulder,
Colorado.
Kingsley, G. R., 1995, “Evaluation and Retrofit of
Unreinforced Masonry Buildings,” Proceedings of the
Third National Concrete and Masonry Engineering
Conference, San Francisco, California, pp. 709–728.
Klingner, R. E., editor, 1994, Performance of Masonry
Structures in the Northridge Earthquake of January 17,
1994, The Masonry Society, Boulder, Colorado.
Klingner, R. E., and Bertero, V. V., 1976, Infilled
Frames in Earthquake Resistant Construction, Report
No. EERC 76-32, Earthquake Engineering Research
Center, University of California at Berkeley, California.
Leiva, G., and Klingner, R. E., 1991, In-Plane Seismic
Resistance of Two-Story Concrete Masonry Shear Walls
FEMA 274 Seismic Rehabilitation Commentary
with Openings, TCCMAR Report No. 3.1(c)-2,
Department of Civil Engineering, University of Texas at
Austin, Texas.
Lenczner, D., 1986, “Creep and Prestress Losses in
Brick Masonry,” The Structural Engineer, Institution of
Structural Engineers, United Kingdom, Vol. 64B, No. 3,
pp. 57–62.
Lotfi, H. R., and Shing, P. B., 1994, “Interface Model
Applied to Fracture of Masonry Structures,” Journal of
the Structural Engineering Division, American Society
of Civil Engineers, New York, New York, Vol. 120,
No. 1.
Mainstone, R. J., and Weeks, G. A., 1971, “The
Influence of a Bounding Frame on the Racking
Stiffness and Strength of Brick Walls,” Current Paper
CP 3/72, Building Research Station, Garston, United
Kingdom, also published in Proceedings of Second
International Brick Masonry Conference, pp. 165–171.
Mainstone, R. J., 1971, “On the Stiffnesses and
Strengths of Infilled Frames,” Current Paper CP 2/72,
Building Research Station, Garston, United Kingdom,
reprinted from Proceedings of Institution of Civil
Engineers, 1971 Supplement (iv), Paper 7360 S, pp.
57–90.
Mainstone, R. J., 1974, “Supplementary Note on the
Stiffnesses and Strengths of Infilled Frames,” Current
Paper CP 13/74, Building Research Station, Garston,
United Kingdom.
Masonry Standards Joint Committee (MSJC), 1995a,
Building Code Requirements for Masonry Structures,
ACI 530-95/ASCE 5-95/TMS 402-95, American
Concrete Institute, Detroit, Michigan, American
Society of Civil Engineers, New York, New York, and
The Masonry Society, Boulder, Colorado.
Masonry Standards Joint Committee (MSJC), 1995b,
Specification for Masonry Structures, ACI 530.1-95/
ASCE 6-95/TMS 602-95, American Concrete Institute,
Detroit, Michigan, American Society of Civil
Engineers, New York, New York, and The Masonry
Society, Boulder, Colorado.
Mehrabi, A. B., and Shing, P. B., 1994, “Performance of
Masonry-Infilled R/C Frames Under In-Plane Lateral
Loads: Analytical Modeling,” Proceedings of NCEER
Workshop on Seismic Response of Masonry Infills, San
7-37
 Chapter 7: Masonry (Systematic Rehabilitation)
Francisco, California, Technical Report
NCEER-94-0004, National Center for Earthquake
Engineering Research, State University of New York at
Buffalo, New York.
Merryman, K. M., Leiva, G., Antrobus, N., and
Klingner, R. E., 1990, In-Plane Seismic Resistance of
Two-Story Concrete Masonry Coupled Shear Walls,
TCCMAR Report No. 3.1(c)-1, Department of Civil
Engineering, University of Texas at Austin, Texas.
NBS, Evaluation of Structural Properties of Masonry in
Existing Buildings, Building Science Series 62,
National Bureau of Standards (now National Institute
for Science and Technology), U.S. Department of
Commerce, Washington, D.C.
Noland, J. L., Atkinson, R. H., and Kingsley, G. R.,
1987, “Nondestructive Methods for Evaluating
Masonry Structures,” Proceedings of International
Conference on Structural Faults and Repair, London,
United Kingdom.
Paulay, T., and Priestley, M.J.N., 1992, Seismic Design
of Reinforced Concrete and Masonry Buildings, Wiley
Interscience, New York, New York.
Paulson, T. J. and Abrams, D. P., 1990, Measured
Inelastic Response of Reinforced Masonry Building
Structures to Earthquake Motions, TCCMAR Report
No. 7.1-1, Structural Research Series No. 555,
University of Illinois at Urbana-Champaign, Illinois.
Plummer, H. C. and Blume, J. A., 1953, Reinforced
Brick Masonry and Lateral Force Design, Structural
Clay Products Institute (now Brick Institute of
America), Reston, VA.
Polyakov, S. V., 1957, “Masonry in Framed Buildings:
An Investigation into the Strength and Stiffness of
Masonry Infilling” (English translation), Moscow,
Russia.
Polyakov, S. V., 1960, “On the Interaction Between
Masonry Filler Walls and Enclosing Frame When
Loaded in the Plane of the Wall,” Translations in
Earthquake Engineering, Earthquake Engineering
Research Institute, Oakland, California, pp. 36–42.
Priestley, M. J. N., and Hart, G. C., 1989, Design
Recommendations for the Period of Vibration of
7-38 Seismic Rehabilitation Commentary
Masonry Wall Buildings, Report No. SSRP-89/05,
Department of Applied Mechanics and Engineering
Sciences, University of California at San Diego,
California.
Sansalone, M., and Carino, N., 1988, “Impact-Echo
Method, Detecting Honeycombing, the Depth of
Surface Opening Cracks, and Ungrouted Tendons,”
Concrete International, American Concrete Institute,
Detroit, Michigan, p. 38.
Seible, F., Hegemier, G. A., Priestley, M.J.N., Kingsley,
G. R., Kurkchubasche, A. G., Igarashi, A., and Weeks,
J. S., 1994, The US-TCCMAR Full-Scale Five-Story
Masonry Research Building Test, Part I - Executive
Summary, TCCMAR Report No. 9.4-1, Department of
Applied Mechanics and Engineering Sciences,
University of California at San Diego, California.
Shing, P. B., Noland, J. L., Spaeh, H. P.,
Klamerus, E. W., and Schuller, M. P., 1991, Response of
Single-Story Reinforced Masonry Shear Walls to In-
Plane Lateral Loads, TCCMAR Report No. 3.1(a)-2,
Department of Civil, Environmental and Architectural
Engineering, University of Colorado at Boulder,
Colorado.
Stafford Smith, B., 1962, “Lateral Stiffness of Infilled
Frames,” Journal of the Structural Engineering
Division, American Society of Civil Engineers, New
York, New York, pp. 183–199.
Stafford Smith, B., 1966, “Behavior of Square Infilled
Frames,” Journal of the Structural Engineering
Division, American Society of Civil Engineers, New
York, New York, pp. 381–403.
Stafford Smith, B., 1968, “Model Test Results of
Vertical and Horizontal Loading of Infilled Frames,”
ACI Structural Journal, American Concrete Institute,
Detroit, Michigan, pp. 618–624.
Stafford Smith, B., and Carter, C., 1969, “A Method of
Analysis for Infilled Frames,” Proceedings of the
Institution of Civil Engineers, London, United
Kingdom, Vol. 44, pp. 31–48.
Tena-Colunga, A., and Abrams, D. P., 1992, Response
of an Unreinforced Masonry Building During the Loma
Prieta Earthquake, Structural Research Series No. 576,
University of Illinois at Urbana-Champaign, Illinois.
FEMA 274
 Chapter 7: Masonry (Systematic Rehabilitation)

Turnsek, V., and Sheppard, P., 1980, “The Shear and
Flexural Resistance of Masonry Walls,” Proceedings of
the International Research Conference on Earthquake
Engineering, Skopje, Yugoslavia.
Xu, W., and Abrams, D. P., 1992, Evaluation of Lateral
Strength and Deflection for Cracked Unreinforced
Masonry Walls, Advanced Construction Technology
Center Report #92-26-11, College of Engineering,
University of Illinois at Urbana, Illinois.

FEMA 274 Seismic Rehabilitation Commentary 7-39

 Chapter 7: Masonry (Systematic Rehabilitation)

7-40 Seismic Rehabilitation Commentary FEMA 274